-
Mancisidor, Rogelio Andrade; Jenssen, Robert, Yu, Shujian & Kampffmeyer, Michael
(2025)
Aggregation of Dependent Expert Distributions in Multimodal Variational Autoencoders
Proceedings of Machine Learning Research (PMLR).
-
Grassi, Stefano; Ravazzolo, Francesco, Vespignani, Joaquin & Vocalelli, Giorgio
(2025)
Global money supply and energy and non-energy commodity prices: A MS-TV-VAR approach
-
Paredes, Rodrigo; Yang, Wei-Ting & Reis, Marco S.
(2025)
Decentralized causal-based monitoring for large-scale systems: sensitivity and robustness assessment
Vis sammendrag
Ensuring safety and efficiency in industrial systems requires effective fault detection and diagnosis, which becomes increasingly challenging in high-dimensional and complex environments. Traditional multivariate statistical process monitoring methods, such as those based on Principal Component Analysis and Partial Least Squares, often fall short in their ability to diagnose localized faults due to their lack of causal modeling. This paper introduces a Causal Network-based Decentralized Multivariate Statistical Process Control (CNd-MSPC) framework, which employs causal networks and community detection—specifically the Leiden algorithm—to segment large systems into functional communities and perform distributed monitoring. This structural partitioning preserves essential causal and topological information, enhancing the sensitivity for fault detection in high-dimensional systems by allowing focused analysis of specific sub-networks. Through extensive testing with a graph-based data simulator, we demonstrate that CNd-MSPC consistently outperforms centralized methods across various network sizes, achieving higher fault detection sensitivity for both process perturbations and sensor biases, especially in large networks. The decentralized approach retains high sensitivity, even when data from several communities are missing due to process disruptions.
-
Fatima, Safia; Moonen, Leon & Ellefsen, Kai Olav
(2025)
Self Healing of a Mixed Autonomy Traffic System Using Reinforcement Learning and Attention
IEEE Open Journal of Intelligent Transportation Systems.
Vis sammendrag
As urban traffic becomes increasingly complex with the integration of connected and autonomous vehicles alongside human-driven vehicles, there is a critical need for adaptive traffic management systems capable of self-healing in response to disruptions. This paper introduces TS2RLA (“Traffic System Recovery using Reinforcement Learning and Attention”), a novel framework for self-healing in mixed-autonomy traffic systems by combining deep reinforcement learning with an attention mechanism to optimize traffic flow and recover from faults in various scenarios in a mixed-autonomy traffic environment. We evaluated TS2RLA in four complex traffic scenarios: bottleneck, figure-eight, grid, and merge. Our results demonstrate significant improvements over the baseline model, showing an average of 86.74% reduction in crashes, 71% improvement in speed and traffic throughput, and robust performance under diverse and complex traffic conditions. Moreover, our experiments show that TS2RLA leads to a significant reduction in CO2 emissions and fuel consumption. TS2RLA’s attention-based approach shows particular benefits in bottleneck and figure-eight scenarios, demonstrating its ability to adapt to complex, multi-factor traffic situations. For scenarios that TS2RLA had not been trained on before, it performs even more favorably than the baseline, with a 96.8% crash reduction and 95.3% throughput improvement. This shows its ability to adapt effectively to new traffic conditions. Overall, we conclude that TS2RLA could significantly improve the safety, efficiency, and capacity of real-world traffic systems, particularly in dynamic urban environments. As such, our work contributes to the field of intelligent transportation systems by offering a versatile self-healing framework capable of managing the complexities of mixed-autonomy traffic.
-
Ovanger, Oscar; Lee, Daesoo, Eidsvik, Jo, Hauge, Ragnar, Skauvold, Jacob & Aune, Erlend
(2025)
A Statistical Study of Latent Diffusion Models for Geological Facies Modeling
-
Lee, Daesoo; Ovanger, Oscar, Eidsvik, Jo, Aune, Erlend, Skauvold, Jacob & Hauge, Ragnar
(2025)
Latent diffusion model for conditional reservoir facies generation
-
Ravazzolo, Francesco & Rossini, Luca
(2025)
IS THE PRICE CAP FOR GAS USEFUL? EVIDENCE FROM EUROPEAN COUNTRIES
-
Goracci, Greta; Ferrari, Davide, Giannerini, Simone & Ravazzolo, Francesco
(2024)
Robust Estimation for Threshold Autoregressive Moving-Average Models
Vis sammendrag
Threshold autoregressive moving-average (TARMA) models extend the popular TAR model and are among the few parametric time series specifications to include a moving average in a nonlinear setting. The state dependent reactions to shocks is particularly appealing in Economics and Finance. However, no theory is currently available when the data present heavy tails or anomalous observations. Here we provide the first theoretical framework for robust M-estimation for TARMA models and study its practical relevance. Under mild conditions, we show that the robust estimator for the threshold parameter is super-consistent, while the estimators for autoregressive and moving-average parameters are strongly consistent and asymptotically normal. The Monte Carlo study shows that the M-estimator is superior, in terms of both bias and variance, to the least squares estimator, which can be heavily affected by outliers. The findings suggest that robust M-estimation should be generally preferred to the least squares method. We apply our methodology to a set of commodity price time series; the robust TARMA fit presents smaller standard errors and superior forecasting accuracy. The results support the hypothesis of a two-regime non-linearity characterized by slow expansions and fast contractions.
-
Lee, Daesoo; Malacarne, Sara & Aune, Erlend
(2024)
Explainable time series anomaly detection using masked latent generative modeling
Vis sammendrag
We present a novel time series anomaly detection method that achieves excellent detection accuracy while offering a superior level of explainability. Our proposed method, TimeVQVAE-AD, leverages masked generative modeling adapted from the cutting-edge time series generation method known as TimeVQVAE. The prior model is trained on the discrete latent space of a time–frequency domain. Notably, the dimensional semantics of the time–frequency domain are preserved in the latent space, enabling us to compute anomaly scores across different frequency bands, which provides a better insight into the detected anomalies. Additionally, the generative nature of the prior model allows for sampling likely normal states for detected anomalies, enhancing the explainability of the detected anomalies through counterfactuals. Our experimental evaluation on the UCR Time Series Anomaly archive demonstrates that TimeVQVAE-AD significantly surpasses the existing methods in terms of detection accuracy and explainability. We provide our implementation on GitHub: https://github.com/ML4ITS/TimeVQVAE-AnomalyDetection.
-
Hoesch, Lukas; Lee, Adam & Mesters, Geert
(2024)
Locally Robust Inference for Non-Gaussian SVAR models
Vis sammendrag
All parameters in structural vector autoregressive (SVAR) models are locally identified when the structural shocks are independent and follow non‐Gaussian distributions. Unfortunately, standard inference methods that exploit such features of the data for identification fail to yield correct coverage for structural functions of the model parameters when deviations from Gaussianity are small. To this extent, we propose a locally robust semiparametric approach to conduct hypothesis tests and construct confidence sets for structural functions in SVAR models. The methodology fully exploits non‐Gaussianity when it is present, but yields correct size/coverage for local‐to‐Gaussian densities. Empirically, we revisit two macroeconomic SVAR studies where we document mixed results. For the oil price model of Kilian and Murphy (2012), we find that non‐Gaussianity can robustly identify reasonable confidence sets, whereas for the labor supply–demand model of Baumeister and Hamilton (2015) this is not the case. Moreover, these exercises highlight the importance of using weak identification robust methods to assess estimation uncertainty when using non‐Gaussianity for identification.
-
Yang, Wei-Ting; Tamssaouet, Karim & Dauzère-Pérès, Stéphane
(2024)
Bayesian network structure learning using scatter search
Vis sammendrag
Learning the structure of Bayesian networks (BNs) from data is an NP-hard problem as the solution space grows super-exponentially with the number of nodes. Many algorithms have been developed to efficiently find the best structures, with score-based algorithms being those that use heuristics or metaheuristics to explore potential structures in the search space. This paper proposes a new score-based algorithm that relies on a well-known metaheuristic called scatter search, which, to the best of our knowledge, has not been used in learning BN structure. The core of scatter search is to maintain a reference set that stores both high-quality and diverse solutions, thereby continuously tracking and improving the high-quality solutions, while exploring different search directions indicated by the diverse solutions. By incorporating a distance metric in the learning process, the exploration can be more systematic than purely random, as is often the case in most existing algorithms. The effectiveness and efficiency of the proposed algorithm are evaluated through computational experiments. In addition to learning higher-score structures, the results show that scatter search provides a higher degree of robustness compared with benchmark algorithms
-
Ivanovska, Magdalena & Slavkovik, Marija
(2024)
Probabilistic judgment aggregation with conditional independence constraints
Vis sammendrag
Probabilistic judgment aggregation is concerned with aggregating judgments about probabilities of logically related issues. It takes as input imprecise probabilistic judgments over the issues given by a group of agents and defines rules of aggregating the individual judgments into a collective opinion representative for the group. The process of aggregation can be subject to constraints, i.e., aggregation rules can be required to satisfy certain properties. We explore how probabilistic independence constraints can be represented and incorporated into the aggregation process.
-
Dreyer, Lars Willas; Eklund, Anders, Rognes, Marie Elisabeth, Malm, Jan, Qvarlander, Sara, Støverud, Karen-Helene, Mardal, Kent-Andre & Vinje, Vegard
(2024)
Modeling CSF circulation and the glymphatic system during infusion using subject specific intracranial pressures and brain geometries
Vis sammendrag
Background Infusion testing is an established method for assessing CSF resistance in patients with idiopathic normal pressure hydrocephalus (iNPH). To what extent the increased resistance is related to the glymphatic system is an open question. Here we introduce a computational model that includes the glymphatic system and enables us to determine the importance of (1) brain geometry, (2) intracranial pressure, and (3) physiological parameters on the outcome of and response to an infusion test. Methods We implemented a seven-compartment multiple network porous medium model with subject specific geometries from MR images using the finite element library FEniCS. The model consists of the arterial, capillary and venous blood vessels, their corresponding perivascular spaces, and the extracellular space (ECS). Both subject specific brain geometries and subject specific infusion tests were used in the modeling of both healthy adults and iNPH patients. Furthermore, we performed a systematic study of the effect of variations in model parameters. Results Both the iNPH group and the control group reached a similar steady state solution when subject specific geometries under identical boundary conditions was used in simulation. The difference in terms of average fluid pressure and velocity between the iNPH and control groups, was found to be less than 6% during all stages of infusion in all compartments. With subject specific boundary conditions, the largest computed difference was a 75% greater fluid speed in the arterial perivascular space (PVS) in the iNPH group compared to the control group. Changes to material parameters changed fluid speeds by several orders of magnitude in some scenarios. A considerable amount of the CSF pass through the glymphatic pathway in our models during infusion, i.e., 28% and 38% in the healthy and iNPH patients, respectively. Conclusions Using computational models, we have found the relative importance of subject specific geometries to be less important than individual differences in resistance as measured with infusion tests and model parameters such as permeability, in determining the computed pressure and flow during infusion. Model parameters are uncertain, but certain variations have large impact on the simulation results. The computations resulted in a considerable amount of the infused volume passing through the brain either through the perivascular spaces or the extracellular space.
-
Astekin, Merve; Hort, Max & Moonen, Leon
(2024)
An Exploratory Study on How Non-Determinism in Large Language Models Affects Log Parsing
Vis sammendrag
Most software systems used in production generate system logs that provide a rich source of information about the status and execution behavior of the system. These logs are commonly used to ensure the reliability and maintainability of software systems. The first step toward automated log analysis is generally log parsing, which aims to transform unstructured log messages into structured log templates and extract the corresponding parameters.
Recently, Large Language Models (LLMs) such as ChatGPT have shown promising results on a wide range of software engineering tasks, including log parsing. However, the extent to which non-determinism influences log parsing using LLMs remains unclear. In particular, it is important to investigate whether LLMs behave consistently when faced with the same log message multiple times.
In this study, we investigate the impact of non-determinism in state-of-the-art LLMs while performing log parsing. Specifically, we select six LLMs, including both paid proprietary and free-to-use models, and evaluate their non-determinism on 16 system logs obtained from a selection of mature open-source projects. The results of our study reveal varying degrees of non-determinism among models. Moreover, they show that there is no guarantee for deterministic results even with a temperature of zero.
-
Vidziunas, Linas; Binkley, David & Moonen, Leon
(2024)
The Impact of Program Reduction on Automated Program Repair
-
Moss, Jonas
(2024)
Measures of Agreement with Multiple Raters: Fréchet Variances and Inference
Vis sammendrag
Most measures of agreement are chance-corrected. They differ in three dimensions: their definition of chance agreement, their choice of disagreement function, and how they handle multiple raters. Chance agreement is usually defined in a pairwise manner, following either Cohen’s kappa or Fleiss’s kappa. The disagreement function is usually a nominal, quadratic, or absolute value function. But how to handle multiple raters is contentious, with the main contenders being Fleiss’s kappa, Conger’s kappa, and Hubert’s kappa, the variant of Fleiss’s kappa where agreement is said to occur only if every rater agrees. More generally, multi-rater agreement coefficients can be defined in a g-wise way, where the disagreement weighting function uses g raters instead of two. This paper contains two main contributions. (a) We propose using Fréchet variances to handle the case of multiple raters. The Fréchet variances are intuitive disagreement measures and turn out to generalize the nominal, quadratic, and absolute value functions to the case of more than two raters. (b) We derive the limit theory of g-wise weighted agreement coefficients, with chance agreement of the Cohen-type or Fleiss-type, for the case where every item is rated by the same number of raters. Trying out three confidence interval constructions, we end up recommending calculating confidence intervals using the arcsine transform or the Fisher transform.
-
Gnoatto, Alessandro; Lavagnini, Silvia & Picarelli, Athena
(2024)
Deep Quadratic Hedging
Vis sammendrag
We propose a novel computational procedure for quadratic hedging in high-dimensional incomplete markets, covering mean-variance hedging and local risk minimization. Starting from the observation that both quadratic approaches can be treated from the point of view of backward stochastic differential equations (BSDEs), we (recursively) apply a deep learning-based BSDE solver to compute the entire optimal hedging strategies paths. This allows us to overcome the curse of dimensionality, extending the scope of applicability of quadratic hedging in high dimension. We test our approach with a classic Heston model and with a multiasset and multifactor generalization thereof, showing that this leads to high levels of accuracy.
-
Lundén, Daniel; Hummelgren, Lars, Kudlicka, Jan, Eriksson, Oscar & Broman, David
(2024)
Suspension Analysis and Selective Continuation-Passing Style for Universal Probabilistic Programming Languages
Vis sammendrag
Universal probabilistic programming languages (PPLs) make it relatively easy to encode and automatically solve statistical inference problems. To solve inference problems, PPL implementations often apply Monte Carlo inference algorithms that rely on execution suspension. State-of-the-art solutions enable execution suspension either through (i) continuation-passing style (CPS) transformations or (ii) efficient, but comparatively complex, low-level solutions that are often not available in high-level languages. CPS transformations introduce overhead due to unnecessary closure allocations—a problem the PPL community has generally overlooked. To reduce overhead, we develop a new efficient selective CPS approach for PPLs. Specifically, we design a novel static suspension analysis technique that determines parts of programs that require suspension, given a particular inference algorithm. The analysis allows selectively CPS transforming the program only where necessary. We formally prove the correctness of the analysis and implement the analysis and transformation in the Miking CorePPL compiler. We evaluate the implementation for a large number of Monte Carlo inference algorithms on real-world models from phylogenetics, epidemiology, and topic modeling. The evaluation results demonstrate significant improvements across all models and inference algorithms.
-
Lee, Adam & Mesters, Geert
(2024)
Locally robust inference for non-Gaussian linear simultaneous equations models
Vis sammendrag
All parameters in linear simultaneous equations models can be identified (up to permutation and sign) if the underlying structural shocks are independent and at most one of them is Gaussian. Unfortunately, existing inference methods that exploit such identifying assumptions suffer from size distortions when the true distributions of the shocks are close to Gaussian. To address this weak non-Gaussian problem we develop a locally robust semi-parametric inference method which is simple to implement, improves coverage and retains good power properties. The finite sample properties of the methodology are illustrated in a large simulation study and an empirical study for the returns to schooling.
-
Foldnes, Njål; Moss, Jonas & Grønneberg, Steffen
(2024)
Improved Goodness of Fit Procedures for Structural Equation Models
Vis sammendrag
We propose new ways of robustifying goodness-of-fit tests for structural equation modeling under non-normality. These test statistics have limit distributions characterized by eigenvalues whose estimates are highly unstable and biased in known directions. To take this into account, we design model-based trend predictions to approximate the population eigenvalues. We evaluate the new procedures in a large-scale simulation study with three confirmatory factor models of varying size (10, 20, or 40 manifest variables) and six non-normal data conditions. The eigenvalues in each simulated dataset are available in a database. Some of the new procedures markedly outperform presently available methods. We demonstrate how the new tests are calculated with a new R package and provide practical recommendations.
-
Schroeder, Daniel Thilo; Orgeret, Kristin Skare, Bruijn, Mirjam de, Bruls, Luca, Moges, Mulatu Alemayehu, Badji, Samba Dialimpa, Fritz, Noémie, Cisse, Modibo Galy, Langguth, Johannes & Mutsvairo, Bruce
(2023)
Social media in the Global South: A Network Dataset of the Malian Twittersphere
-
Brenner, Stefan; Schroeder, Daniel Thilo & Langguth, Johannes
(2023)
GECO: A Twitter Dataset of COVID-19 Misinformation and Conspiracy Theories Related to the Berlin Parliament and Washington Capitol Riots
NIKT: Norsk IKT-konferanse for forskning og utdanning.
-
Juelsrud, Ragnar Enger & Larsen, Vegard Høghaug
(2023)
Macroeconomic uncertainty and bank lending
Vis sammendrag
We investigate the impact of macro-related uncertainty on bank lending in Norway. We show that an increase in general macroeconomic uncertainty reduces bank lending. Importantly, however, we show that this effect is largely driven by monetary policy uncertainty, suggesting that uncertainty about the monetary policy stance is key for understanding why macro-related uncertainty impacts bank lending.
-
Casarin, Roberto; Grassi, Stefano, Ravazzolo, Francesco & Dijk, Herman K. van
(2023)
A flexible predictive density combination for large financial data sets in regular and crisis periods
Vis sammendrag
A flexible predictive density combination is introduced for large financial data sets which allows for model set incompleteness. Dimension reduction procedures that include learning allocate the large sets of predictive densities and combination weights to relatively small subsets. Given the representation of the probability model in extended nonlinear state-space form, efficient simulation-based Bayesian inference is proposed using parallel dynamic clustering as well as nonlinear filtering, implemented on graphics processing units. The approach is applied to combine predictive densities based on a large number of individual US stock returns of daily observations over a period that includes the Covid-19 crisis period. Evidence on dynamic cluster composition, weight patterns and model set incompleteness gives valuable signals for improved modelling. This enables higher predictive accuracy and better assessment of uncertainty and risk for investment fund management.
-
Mancisidor, Rogelio Andrade; Kampffmeyer, Michael Christian, Aas, Kjersti & Jenssen, Robert
(2023)
Discriminative multimodal learning via conditional priors in generative models
Vis sammendrag
Deep generative models with latent variables have been used lately to learn joint representations and generative processes from multi-modal data, which depict an object from different viewpoints. These two learning mechanisms can, however, conflict with each other and representations can fail to embed information on the data modalities. This research studies the realistic scenario in which all modalities and class labels are available for model training, e.g. images or handwriting, but where some modalities and labels required for downstream tasks are missing, e.g. text or annotations. We show, in this scenario, that the variational lower bound limits mutual information between joint representations and missing modalities. We, to counteract these problems, introduce a novel conditional multi-modal discriminative model that uses an informative prior distribution and optimizes a likelihood-free objective function that maximizes mutual information between joint representations and missing modalities. Extensive experimentation demonstrates the benefits of our proposed model, empirical results show that our model achieves state-of-the-art results in representative problems such as downstream classification, acoustic inversion, and image and annotation generation.
-
Asimakopoulos, Stylianos; Lorusso, Marco & Ravazzolo, Francesco
(2023)
A Bayesian DSGE approach to modelling cryptocurrency
Vis sammendrag
We develop and estimate a DSGE model to evaluate the economic repercussions of cryptocurrency. In our model, cryptocurrency offers an alternative currency option to government currency, with endogenous supply and demand. We uncover a substitution effect between the real balances of government currency and cryptocurrency in response to technology, preferences and monetary policy shocks. We find that an increase in cryptocurrency productivity induces a rise in the relative price of government currency with respect to cryptocurrency. Since cryptocurrency and government currency are highly substitutable, the demand for the former increases whereas it drops for the latter. Our historical decomposition analysis shows that fluctuations in the cryptocurrency price are mainly driven by shocks in cryptocurrency demand, whereas changes in the real balances for government currency are mainly attributed to government currency and cryptocurrency demand shocks.
-
Behmiri, Niaz Bashiri; Fezzi, Carlo & Ravazzolo, Francesco
(2023)
Incorporating air temperature into mid-term electricity load forecasting models using time-series regressions and neural networks
Vis sammendrag
One of the most controversial issues in the mid-term load forecasting literature is the treatment of weather. Because of the difficulty in obtaining precise weather forecasts for a few weeks ahead, researchers have, so far, implemented three approaches: a) excluding weather from load forecasting models altogether, b) assuming future weather to be perfectly known and c) including weather forecasts in their load forecasting models. This article provides the first systematic comparison of how the different treatments of weather affect load forecasting performance. We incorporate air temperature into short- and mid-term load forecasting models, comparing time-series methods and feed-forward neural networks. Our results indicate that models including future temperature always significantly outperform models excluding temperature, at all-time horizons. However, when future temperature is replaced with its prediction, these results become weaker.
-
Høst, Anders Mølmen; Lison, Pierre & Moonen, Leon
(2023)
Constructing a Knowledge Graph from Textual Descriptions of Software Vulnerabilities in the National Vulnerability Database
Vis sammendrag
Knowledge graphs have shown promise for several cybersecurity tasks, such as vulnerability assessment and threat analysis. In this work, we present a new method for constructing a vulnerability knowledge graph from information in the National Vulnerability Database (NVD). Our approach combines named entity recognition (NER), relation extraction (RE), and entity prediction using a combination of neural models, heuristic rules, and knowledge graph embeddings. We demonstrate how our method helps to fix missing entities in
knowledge graphs used for cybersecurity and evaluate the performance.
-
Miroshnychenko, Ivan; Vocalelli, Giorgio, Massis, Alfredo De, Grassi, Stefano & Ravazzolo, Francesco
(2023)
The COVID-19 pandemic and family business performance
Vis sammendrag
This study examines the impact of the COVID-19 pandemic on corporate financial performance using a unique, cross-country, and longitudinal sample of 3350 listed firms worldwide. We find that the financial performance of family firms has been significantly higher than that of nonfamily firms during the COVID-19 pandemic, accounting for pre-pandemic business conditions. This effect is pertinent to firms with strong family involvement in management or in both management and ownership. We also identify the role of firm-, industry-, and country-level contingencies for family business financial performance during the COVID-19 pandemic. This study offers a novel understanding of the financial resilience across different types of family business and sets an agenda for future research on the drivers of resilience of family firms to adverse events. It also provides important and novel evidence for policymakers, particularly for firms with different ownership and management structures.
-
Durante, F.; Gatto, A. & Ravazzolo, Francesco
(2023)
Understanding relationships with the Aggregate Zonal Imbalance using copulas
Vis sammendrag
In the Italian electricity market, we analyze the Aggregate Zonal Imbalance, which is the algebraic sum, changed in sign, of the amount of energy procured by the Italian national Transmission and System Operator in the Dispatching Services Market at a given time in the northern Italian electricity macro-zone. Specifically, we determine possible relationships among the Aggregate Zonal Imbalances and other variables of interest in electricity markets, including renewable sources. From a methodological point of view, we use a multivariate model for time series that combines the marginal behavior with copula-type models. As a result, the flexibility of a copula approach will allow identifying the nature of non-linear linkages among the Aggregate Zonal Imbalance and other variables such as forecasted demand, forecasted wind and solar PV generation. In this respect, novel ways to measure dependence and association among random variates are adopted. Our results indicate a clear association between the Aggregate Zonal Imbalance and Forecasted Solar PV generation, and a weaker relationship with the other considered variables. We find this result both in terms of pairwise Spearman’s and Kendall’s correlations and in terms of upper and lower tail dependence. The analysis concludes with the proposal of new indicators to detect association among random vectors, which could identify the more important features driving imbalances.
-
Haugsdal, Espen; Aune, Erlend & Ruocco, Massimiliano
(2023)
Persistence Initialization: a novel adaptation of the Transformer architecture for time series forecasting
Vis sammendrag
Time series forecasting is an important problem, with many real world applications. Transformer models have been successfully applied to natural language processing tasks, but have received relatively little attention for time series forecasting. Motivated by the differences between classification tasks and forecasting, we propose PI-Transformer, an adaptation of the Transformer architecture designed for time series forecasting, consisting of three parts: First, we propose a novel initialization method called Persistence Initialization, with the goal of increasing training stability of forecasting models by ensuring that the initial outputs of an untrained model are identical to the outputs of a simple baseline model. Second, we use ReZero normalization instead of Layer Normalization, in order to further tackle issues related to training stability. Third, we use Rotary positional encodings to provide a better inductive bias for forecasting. Multiple ablation studies show that the PI-Transformer is more accurate, learns faster, and scales better than regular Transformer models. Finally, PI-Transformer achieves competitive performance on the challenging M4 dataset, both when compared to the current state of the art, and to recently proposed Transformer models for time series forecasting.
-
Al-Bataineh, Omar; Moonen, Leon & Vidziunas, Linas
(2023)
Extending the range of bugs that automated program repair can handle
Vis sammendrag
Modern automated program repair (APR) is well-tuned to finding and repairing bugs that introduce observable erroneous behavior to a program. However, a significant class of bugs does not lead to observable behavior (e.g., termination bugs and non-functional bugs). Such bugs can generally not be handled with current APR approaches, so complementary techniques are needed. To stimulate the systematic study of alternative approaches and hybrid combinations, we devise a novel bug classification system that enables methodical analysis of their bug detection power and bug repair capabilities. To demonstrate the benefits, we study the repair of termination bugs in sequential and concurrent programs. Our analysis shows that integrating dynamic APR with formal analysis techniques, such as termination provers and software model checkers, reduces complexity and improves the overall reliability of these repairs. We empirically investigate how well the hybrid approach can repair termination and performance bugs by experimenting with hybrids that integrate different APR approaches with termination provers and execution time monitors. Our findings indicate that hybrid repair holds promise for handling termination and performance bugs. However, the capability of the chosen tools and the completeness of the available correctness specification affects the quality of the patches that can be produced.
-
Lee, Daesoo; Malacarne, Sara & Aune, Erlend
(2023)
Vector Quantized Time Series Generation with a Bidirectional Prior Model
Proceedings of Machine Learning Research (PMLR), 206, p. 7665-7693.
Vis sammendrag
Time series generation (TSG) studies have
mainly focused on the use of Generative Adversarial Networks (GANs) combined with recurrent neural network (RNN) variants. However, the fundamental limitations and challenges
of training GANs still remain. In addition,
the RNN-family typically has difficulties with
temporal consistency between distant timesteps.
Motivated by the successes in the image generation (IMG) domain, we propose TimeVQVAE,
the first work, to our knowledge, that uses vector quantization (VQ) techniques to address the
TSG problem. Moreover, the priors of the discrete latent spaces are learned with bidirectional
transformer models that can better capture global
temporal consistency. We also propose VQ modeling in a time-frequency domain, separated into
low-frequency (LF) and high-frequency (HF).
This allows us to retain important characteristics of the time series and, in turn, generate
new synthetic signals that are of better quality, with sharper changes in modularity, than
its competing TSG methods. Our experimental evaluation is conducted on all datasets from
the UCR archive, using well-established metrics
in the IMG literature, such as Frechet inception ´
distance and inception scores.
-
López, Ovielt Baltodano; Bulfone, Giacomo, Casarin, Roberto & Ravazzolo, Francesco
(2023)
Modeling Corporate CDS Spreads Using Markov Switching Regressions
Vis sammendrag
This paper investigates the determinants of the European iTraxx corporate CDS index considering a large set of explanatory variables within a Markov switching model framework. The influence of financial and economic variables on CDS spreads are compared using linear, two, three, and four-regime models in a sample post-subprime financial crisis up to the COVID-19 pandemic. Results indicate that four regimes are necessary to model the CDS spreads. The fourth regime was activated during the COVID-19 pandemic and in high volatility periods. Further, the effect of the covariates differs significantly across regimes. Brent and term structure factors became relevant after the outbreak of the COVID-19 pandemic.
-
Foroni, Claudia; Ravazzolo, Francesco & Rossini, Luca
(2023)
Are low frequency macroeconomic variables important for high frequency electricity prices?
Vis sammendrag
Recent research finds that forecasting electricity prices is very relevant. In many applications, it might be interesting to predict daily electricity prices by using their own lags or renewable energy sources. However, the recent turmoil of energy prices and the Russian–Ukrainian war increased attention in evaluating the relevance of industrial production and the Purchasing Managers’ Index output survey in forecasting the daily electricity prices. We develop a Bayesian reverse unrestricted MIDAS model which accounts for the mismatch in frequency between the daily prices and the monthly macro variables in Germany and Italy. We find that the inclusion of macroeconomic low frequency variables is more important for short than medium term horizons by means of point and density measures. In particular, accuracy increases by combining hard and soft information, while using only surveys gives less accurate forecasts than using only industrial production data.
-
Liventsev, Vadim; Grishina, Anastasiia, Härmä, Aki & Moonen, Leon
(2023)
Fully Autonomous Programming with Large Language Models
Vis sammendrag
Current approaches to program synthesis with Large Language Models (LLMs) exhibit a “near miss syndrome”: they tend to generate programs that semantically resemble the correct answer (as measured by text similarity metrics or human evaluation), but achieve a low or even zero accuracy as measured by unit tests due to small imperfections, such as the wrong input or output format. This calls for an approach known as Synthesize, Execute, Debug (SED), whereby a draft of the solution is generated first, followed by a program repair phase addressing the failed tests. To effectively apply this approach to instruction-driven LLMs, one needs to determine which prompts perform best as instructions for LLMs, as well as strike a balance between repairing unsuccessful programs and replacing them with newly generated ones. We explore these trade-offs empirically, comparing replace-focused, repair-focused, and hybrid debug strategies, as well as different template-based and model-based prompt-generation techniques. We use OpenAI Codex as the LLM and Program Synthesis Benchmark 2 as a database of problem descriptions and tests for evaluation. The resulting framework outperforms both conventional usage of Codex without the repair phase and traditional genetic programming approaches.
-
Malik, Sehrish; Naqvi, Moeen & Moonen, Leon
(2023)
CHESS: A Framework for Evaluation of Self-adaptive Systems based on Chaos Engineering
Vis sammendrag
There is an increasing need to assess the correct behavior of self-adaptive and self-healing systems due to their adoption in critical and highly dynamic environments. However, there is a lack of systematic evaluation methods for self-adaptive and self-healing systems. We proposed CHESS, a novel approach to address this gap by evaluating self-adaptive and self-healing systems through fault injection based on chaos engineering (CE). The artifact presented in this paper provides an extensive overview of the use of CHESS through two microservice-based case studies: a smart office case study and an existing demo application called Yelb. It comes with a managing system service, a self-monitoring service, as well as five fault injection scenarios covering infrastructure faults and functional faults. Each of these components can be easily extended or replaced to adopt the CHESS approach to a new case study, help explore its promises and limitations, and identify directions for future research.
-
Galdi, Giulio; Casarin, Roberto, Ferrari, Davide, Fezzi, Carlo & Ravazzolo, Francesco
(2023)
Nowcasting industrial production using linear and non-linear models of electricity demand
Vis sammendrag
This article proposes different modelling approaches which exploit electricity market data to nowcast industrial production. Our models include linear, mixed-data sampling (MIDAS), Markov-Switching (MS) and MS-MIDAS regressions. Comparisons against autoregressive approaches and other commonly used macroeconomic predictors show that electricity market data combined with an MS model significantly improve nowcasting performance, especially during turbulent economic states, such as those generated by the recent COVID-19 pandemic. The most promising results are provided by an MS model which identifies two volatility regimes. These results confirm that electricity market data provide timely and easy-to-access information for nowcasting macroeconomic variables, especially when it is most valuable, i.e. during times of crisis and uncertainty.
-
Colladon, Andrea Fronzetti; Grippa, Francesca, Guardabascio, Barbara, Costante, Gabriele & Ravazzolo, Francesco
(2023)
Forecasting consumer confidence through semantic network analysis of online news
Vis sammendrag
This research studies the impact of online news on social and economic consumer perceptions through semantic network analysis. Using over 1.8 million online articles on Italian media covering four years, we calculate the semantic importance of specific economic-related keywords to see if words appearing in the articles could anticipate consumers’ judgments about the economic situation and the Consumer Confidence Index. We use an innovative approach to analyze big textual data, combining methods and tools of text mining and social network analysis. Results show a strong predictive power for the judgments about the current households and national situation. Our indicator offers a complementary approach to estimating consumer confidence, lessening the limitations of traditional survey-based methods.
-
Moss, Jonas & Grønneberg, Steffen
(2023)
Partial Identification of Latent Correlations with Ordinal Data
Vis sammendrag
The polychoric correlation is a popular measure of association for ordinal data. It estimates a latent correlation, i.e., the correlation of a latent vector. This vector is assumed to be bivariate normal, an assumption that cannot always be justified. When bivariate normality does not hold, the polychoric correlation will not necessarily approximate the true latent correlation, even when the observed variables have many categories. We calculate the sets of possible values of the latent correlation when latent bivariate normality is not necessarily true, but at least the latent marginals are known. The resulting sets are called partial identification sets, and are shown to shrink to the true latent correlation as the number of categories increase. Moreover, we investigate partial identification under the additional assumption that the latent copula is symmetric, and calculate the partial identification set when one variable is ordinal and another is continuous. We show that little can be said about latent correlations, unless we have impractically many categories or we know a great deal about the distribution of the latent vector. An open-source R package is available for applying our results.
-
Iwaszkiewicz-Eggebrecht, Elzbieta; Ronquist, Fredrik, Łukasik, Piotr, Granqvist, Emma, Buczek, Mateusz, Prus, Monika, Kudlicka, Jan, Roslin, Tomas, Tack, Ayco J. M., Andersson, Anders F. & Miraldo, Andreia
(2023)
Optimizing insect metabarcoding using replicated mock communities
Vis sammendrag
Metabarcoding (high-throughput sequencing of marker gene amplicons) has emerged as a promising and cost-effective method for characterizing insect community samples. Yet, the methodology varies greatly among studies and its performance has not been systematically evaluated to date. In particular, it is unclear how accurately metabarcoding can resolve species communities in terms of presence-absence, abundance and biomass.
Here we use mock community experiments and a simple probabilistic model to evaluate the effect of different DNA extraction protocols on metabarcoding performance. Specifically, we ask four questions: (Q1) How consistent are the recovered community profiles across replicate mock communities?; (Q2) How does the choice of lysis buffer affect the recovery of the original community?; (Q3) How are community estimates affected by differing lysis times and homogenization? and (Q4) Is it possible to obtain adequate species abundance estimates through the use of biological spike-ins?
We show that estimates are quite variable across community replicates. In general, a mild lysis protocol is better at reconstructing species lists and approximate counts, while homogenization is better at retrieving biomass composition. Small insects are more likely to be detected in lysates, while some tough species require homogenization to be detected. Results are less consistent across biological replicates for lysates than for homogenates. Some species are associated with strong PCR amplification bias, which complicates the reconstruction of species counts. Yet, with adequate spike-in data, species abundance can be determined with roughly 40% standard error for homogenates, and with roughly 50% standard error for lysates, under ideal conditions. In the latter case, however, this often requires species-specific reference data, while spike-in data generalize better across species for homogenates.
We conclude that a non-destructive, mild lysis approach shows the highest promise for the presence/absence description of the community, while also allowing future morphological or molecular work on the material. However, homogenization protocols perform better for characterizing community composition, in particular in terms of biomass.
-
Billé, Anna Gloria; Tomelleri, Alessio & Ravazzolo, Francesco
(2023)
Forecasting regional GDPs: a comparison with spatial dynamic panel data models
Vis sammendrag
The monitoring of the regional (provincial) economic situation is of particular importance due to the high level of heterogeneity and interdependences among different territories. Although econometric models allow for spatial and serial correlation of various kinds, the limited availability of territorial data restricts the set of relevant predictors at a more disaggregated level, especially for gross domestic product (GDP). Combining data from different sources at NUTS-3 level, this paper evaluates the predictive performance of a spatial dynamic panel data model with individual fixed effects and some relevant exogenous regressors, by using data on total gross value added (GVA) for 103 Italian provinces over the period 2000–2016. A comparison with nested panel sub-specifications as well as pure temporal autoregressive specifications has also been included. The main finding is that the spatial dynamic specification increases forecast accuracy more than its competitors throughout the out-of-sample, recognising an important role played by both space and time. However, when temporal cointegration is detected, the random-walk specification is still to be preferred in some cases even in the presence of short panels.
-
Langguth, Johannes; Schroeder, Daniel Thilo, Filkukova, Petra, Brenner, Stefan, Phillips, Jesper & Pogorelov, Konstantin
(2023)
COCO: an annotated Twitter dataset of COVID-19 conspiracy theories
Vis sammendrag
The COVID-19 pandemic has been accompanied by a surge of misinformation on social media which covered a wide range of different topics and contained many competing narratives, including conspiracy theories. To study such conspiracy theories, we created a dataset of 3495 tweets with manual labeling of the stance of each tweet w.r.t. 12 different conspiracy topics. The dataset thus contains almost 42,000 labels, each of which determined by majority among three expert annotators. The dataset was selected from COVID-19 related Twitter data spanning from January 2020 to June 2021 using a list of 54 keywords. The dataset can be used to train machine learning based classifiers for both stance and topic detection, either individually or simultaneously. BERT was used successfully for the combined task. The dataset can also be used to further study the prevalence of different conspiracy narratives. To this end we qualitatively analyze the tweets, discussing the structure of conspiracy narratives that are frequently found in the dataset. Furthermore, we illustrate the interconnection between the conspiracy categories as well as the keywords.
-
Grishina, Anastasiia; Hort, Max & Moonen, Leon
(2023)
The EarlyBIRD Catches the Bug: On Exploiting Early Layers of Encoder Models for More Efficient Code Classification
Vis sammendrag
The use of modern Natural Language Processing (NLP) techniques has shown to be beneficial for software engineering tasks, such as vulnerability detection and type inference. However, training deep NLP models requires significant computational resources. This paper explores techniques that aim at achieving the best usage of resources and available information in these models. We propose a generic approach, EarlyBIRD, to build composite representations of code from the early layers of a pre-trained transformer model. We empirically investigate the viability of this approach on the CodeBERT model by comparing the performance of 12 strategies for creating composite representations with the standard practice of only using the last encoder layer. Our evaluation on four datasets shows that several early layer combinations yield better performance on defect detection, and some combinations improve multi-class classification. More specifically, we obtain a +2 average improvement of detection accuracy on Devign with only 3 out of 12 layers of CodeBERT and a 3.3x speed-up of fine-tuning. These findings show that early layers can be used to obtain better results using the same resources, as well as to reduce resource usage during fine-tuning and inference.
-
Gåsvær, Kaspara Skovli; Lind, Pedro, Langguth, Johannes, Hjorth-Jensen, Morten, Kreil, Michael & Schroeder, Daniel Thilo
(2023)
Harmful Conspiracies in Temporal Interaction Networks: Understanding the Dynamics of Digital Wildfires through Phase Transitions
-
Lee, Daesoo; Ovanger, Oscar, Eidsvik, Jo, Aune, Erlend, Skauvold, Jacob & Hauge, Ragnar
(2023)
Latent Diffusion Model for Conditional Reservoir Facies Generation
arXiv.
Vis sammendrag
https://arxiv.org/pdf/2311.01968.pdf
-
Moss, Jonas
(2023)
Measuring Agreement Using Guessing Models and Knowledge Coefficients
Vis sammendrag
Several measures of agreement, such as the Perreault–Leigh coefficient, the AC1
, and the recent coefficient of van Oest, are based on explicit models of how judges make their ratings. To handle such measures of agreement under a common umbrella, we propose a class of models called guessing models, which contains most models of how judges make their ratings. Every guessing model have an associated measure of agreement we call the knowledge coefficient. Under certain assumptions on the guessing models, the knowledge coefficient will be equal to the multi-rater Cohen’s kappa, Fleiss’ kappa, the Brennan–Prediger coefficient, or other less-established measures of agreement. We provide several sample estimators of the knowledge coefficient, valid under varying assumptions, and their asymptotic distributions. After a sensitivity analysis and a simulation study of confidence intervals, we find that the Brennan–Prediger coefficient typically outperforms the others, with much better coverage under unfavorable circumstances.
-
Larsen, Vegard Høghaug; Maffei-Faccioli, Nicolo & Pagenhardt, Laura
(2023)
Where do they care? The ECB in the media and inflation expectations
Vis sammendrag
This paper examines how news coverage of the European Central Bank (ECB) affects consumer inflation expectations in the four largest euro area countries. Utilizing a unique dataset of multilingual European news articles, we measure the impact of ECB-related inflation news on inflation expectations. Our results indicate that German and Italian consumers are more attentive to this news, whereas in Spain and France, we observe no significant response. The research underscores the role of national media in disseminating ECB messages and the diverse reactions among consumers in different euro area countries.
-
Hort, Max; Grishina, Anastasiia & Moonen, Leon
(2023)
An Exploratory Literature Study on Sharing and Energy Use of Language Models for Source Code
Vis sammendrag
Large language models trained on source code can support a variety of software development tasks, such as code recommendation and program repair. Large amounts of data for training such models benefit the models’ performance. However, the size of the data and models results in long training times and high energy consumption. While publishing source code allows for replicability, users need to repeat the expensive training process if models are not shared. GOALS: The main goal of the study is to investigate if publications that trained language models for software engineering (SE) tasks share source code and trained artifacts. The second goal is to analyze the transparency on training energy usage. Leon Moonen Simula Research Laboratory & BI Norwegian Business School Oslo, Norway leon.moonen@computer.org understanding, video content prediction [3, 4]). In particular, Deep Learning (DL) often achieves performance improvements by increasing the amount of training data and the size of the model, leading to long training times and substantial energy consumption [5], with an increase in computational costs for state-of-the-art models by a factor of 300000 between 2012 and 2018 [6, 7]. This trend not only raises barriers for researchers with limited computational resources [8], it is also harmful to the environment [5, 6]. METHODS: We perform a snowballing-based literature search to find publications on language models for source code, and analyze their reusability from a sustainability standpoint. RESULTS: From a total of 494 unique publications, we identified 293 relevant publications that use language models to address code-related tasks. Among them, 27% (79 out of 293) make artifacts available for reuse. This can be in the form of tools or IDE plugins designed for specific tasks or task-agnostic models that can be fine-tuned for a variety of downstream tasks. Moreover, we collect insights on the hardware used for model training, as well as training time, which together determine the energy consumption of the development process. CONCLUSION: We find that there are deficiencies in the sharing of information and artifacts for current studies on source code models for software engineering tasks, with 40% of the surveyed papers not sharing source code or trained artifacts. We recommend the sharing of source code as well as trained artifacts, to enable sustainable reproducibility. Moreover, comprehensive information on training times and hardware configurations should be shared for transparency on a model’s carbon footprint. Index Terms—sustainability, reuse, replication, energy, DL4SE.
-
Langguth, Johannes; Tumanis, Aigar & Azad, Ariful
(2022)
Incremental Clustering Algorithms for Massive Dynamic Graphs
Vis sammendrag
We consider the problem of incremental graph clustering where the graph to be clustered is given as a sequence of disjoint subsets of the edge set. The problem appears when dealing with graphs that are created over time, such as online social networks where new users appear continuously, or protein interaction networks when new proteins are discovered. For very large graphs, it is computationally too expensive to repeatedly apply standard clustering algorithms. Instead, algorithms whose time complexity only depends on the size of the incoming subset of edges in every step are needed. At the same time, such algorithms should find clusterings whose quality is close to that produced by offline algorithms. In this paper, we discuss the computational model and present an incremental clustering algorithm. We test the algorithm performance and quality on a wide variety of instances. Our results show that the algorithm far outperforms offline algorithms while retaining a large fraction of their clustering quality.
-
Langguth, Johannes; Filkukova, Petra, Brenner, Stefan, Schroeder, Daniel Thilo & Pogorelov, Konstantin
(2022)
COVID-19 and 5G conspiracy theories: long term observation
of a digital wildfire
Vis sammendrag
The COVID-19 pandemic has severely affected the lives of people worldwide, and consequently, it has dominated world news since March 2020. Thus, it is no surprise that it has also been the topic of a massive amount of misinformation, which was most likely amplified by the fact that many details about the virus were not known at the start of the pandemic. While a large amount of this misinformation was harmless, some narratives spread quickly and had a dramatic real-world effect. Such events are called digital wildfires. In this paper we study a specific digital wildfire: the idea that the COVID-19 outbreak is somehow connected to the introduction of 5G wireless technology, which caused real-world harm in April 2020 and beyond. By analyzing early social media contents we investigate the origin of this digital wildfire and the developments that lead to its wide spread. We show how the initial idea was derived from existing opposition to wireless networks, how videos rather than tweets played a crucial role in its propagation, and how commercial interests can partially explain the wide distribution of this particular piece of misinformation. We then illustrate how the initial events in the UK were echoed several months later in different countries around the world.
-
Straume, Hans-Martin; Asche, Frank, Oglend, Atle, Abrahamsen, Eirik Bjorheim, Birkenbach, Anna M., Langguth, Johannes, Lanquepin, Guillaume & Roll, Kristin Helen
(2022)
Impacts of Covid-19 on Norwegian salmon exports: A firm-level analysis
Vis sammendrag
A rapidly growing literature investigates how the recent Covid-19 pandemic has affected international seafood trade along multiple dimensions, creating opportunities as well as challenges. This suggests that many of the impacts of the Covid measures are subtle and require disaggregated data to allow the impacts in different supply chains to be teased out. In aggregate, Norwegian salmon exports have not been significantly impacted by Covid-related measures. Using firm-level data to all export destinations to examine the effects of lockdowns in different destination countries in 2020, we show that the Covid-related lockdown measures significantly impacted trade patterns for four product forms of salmon. The results also illustrate how the Covid measures create opportunities, as increased stringency of the measures increased trade for two of the product forms. We also find significant differences among firms' responses, with large firms with larger trade networks reacting more strongly to the Covid measures. The limited overall impacts and the significant dynamics at the firm level clearly show the resiliency of the salmon supply chains.
-
Colladon, Andrea Fronzetti; Grassi, Stefano, Ravazzolo, Francesco & Violante, Francesco
(2022)
Forecasting financial markets with semantic network analysis in the COVID-19 crisis
Vis sammendrag
This paper uses a new textual data index for predicting stock market data. The index is applied to a large set of news to evaluate the importance of one or more general economic-related keywords appearing in the text. The index assesses the importance of the economic-related keywords, based on their frequency of use and semantic network position. We apply it to the Italian press and construct indices to predict Italian stock and bond market returns and volatilities in a recent sample period, including the COVID-19 crisis. The evidence shows that the index captures the different phases of financial time series well. Moreover, results indicate strong evidence of predictability for bond market data, both returns and volatilities, short and long maturities, and stock market volatility.
-
Stoltenberg, Emil Aas; Mykland, Per A. & Zhang, Lan
(2022)
A CLT FOR SECOND DIFFERENCE ESTIMATORS WITH AN APPLICATION TO VOLATILITY AND INTENSITY
Vis sammendrag
In this paper, we introduce a general method for estimating the quadratic covariation of one or more spot parameter processes associated with continuous time semimartingales, and present a central limit theorem that has this class of estimators as one of its applications. The class of estimators we introduce, that we call Two-Scales Quadratic Covariation (
TSQC
) estimators, is based on sums of increments of second differences of the observed processes, and the intervals over which the differences are computed are rolling and overlapping. This latter feature lets us take full advantage of the data, and, by sufficiency considerations, ought to outperform estimators that are based on only one partition of the observational window. Moreover, a two-scales approach is employed to deal with asymptotic bias terms in a systematic manner, thus automatically giving consistent estimators without having to work out the form of the bias term on a case-to-case basis. We highlight the versatility of our central limit theorem by applying it to a novel leverage effect estimator that does not belong to the class of
TSQC
estimators. The principal empirical motivation for the present study is that the discrete times at which a continuous time semimartingale is observed might depend on features of the observable process other than its level, such as its spot-volatility process. As an application of the
TSQC
estimators, we therefore show how it may be used to estimate the quadratic covariation between the spot-volatility process and the intensity process of the observation times, when both of these are taken to be semimartingales. The finite sample properties of this estimator are studied by way of a simulation experiment, and we also apply this estimator in an empirical analysis of the Apple stock. Our analysis of the Apple stock indicates a rather strong correlation between the spot volatility process of the log-prices process and the times at which this stock is traded and hence observed.
-
Maulana, Asep; Pogorelov, Konstantin, Schroeder, Daniel Thilo & Langguth, Johannes
(2022)
Graph Neural Network for Fake News Detection and Classification of Unlabelled Nodes at MediaEval 2022
CEUR Workshop Proceedings.
-
Hougen, Conrad D.; Kaplan, Lance M., Ivanovska, Magdalena, Cerutti, Federico, Mishra, Kumar Vijay & III, Alfred O. Hero
(2022)
SOLBP: Second-Order Loopy Belief Propagation for Inference in Uncertain Bayesian Networks
Vis sammendrag
In second-order uncertain Bayesian networks, the conditional probabilities are only known within distributions, i.e., probabilities over probabilities. The delta-method has been applied to extend exact first-order inference methods to propagate both means and variances through sum-product networks derived from Bayesian networks, thereby characterizing epistemic uncertainty, or the uncertainty in the model itself. Alternatively, second-order belief propagation has been demonstrated for polytrees but not for general directed acyclic graph structures. In this work, we extend Loopy Belief Propagation to the setting of second-order Bayesian networks, giving rise to Second-Order Loopy Belief Propagation (SOLBP). For second-order Bayesian networks, SOLBP generates inferences consistent with those generated by sum-product networks, while being more computationally efficient and scalable.
-
Iacopini, Matteo; Ravazzolo, Francesco & Rossini, Luca
(2022)
Proper Scoring Rules for Evaluating Density Forecasts with Asymmetric Loss Functions
Vis sammendrag
This article proposes a novel asymmetric continuous probabilistic score (ACPS) for evaluating and comparing density forecasts. It generalizes the proposed score and defines a weighted version, which emphasizes regions of interest, such as the tails or the center of a variable’s range. The (weighted) ACPS extends the symmetric (weighted) CRPS by allowing for asymmetries in the preferences underlying the scoring rule. A test is used to statistically compare the predictive ability of different forecasts. The ACPS is of general use in any situation where the decision-maker has asymmetric preferences in the evaluation of the forecasts. In an artificial experiment, the implications of varying the level of asymmetry in the ACPS are illustrated. Then, the proposed score and test are applied to assess and compare density forecasts of macroeconomic relevant datasets (U.S. employment growth) and of commodity prices (oil and electricity prices) with particular focus on the recent COVID-19 crisis period.
-
Mancisidor, Rogelio Andrade; Kampffmeyer, Michael, Aas, Kjersti & Jenssen, Robert
(2022)
Generating customer's credit behavior with deep generative models
Vis sammendrag
Innovation is considered essential for today's organizations to survive and thrive. Researchers have also stressed the importance of leadership as a driver of followers' innovative work behavior (FIB). Yet, despite a large amount of research, three areas remain understudied: (a) The relative importance of different forms of leadership for FIB; (b) the mechanisms through which leadership impacts FIB; and (c) the degree to which relationships between leadership and FIB are generalizable across cultures. To address these lacunae, we propose an integrated model connecting four types of positive leadership behaviors, two types of identification (as mediating variables), and FIB. We tested our model in a global data set comprising responses of N = 7,225 participants from 23 countries, grouped into nine cultural clusters. Our results indicate that perceived LMX quality was the strongest relative predictor of FIB. Furthermore, the relationships between both perceived LMX quality and identity leadership with FIB were mediated by social identification. The indirect effect of LMX on FIB via social identification was stable across clusters, whereas the indirect effects of the other forms of leadership on FIB via social identification were stronger in countries high versus low on collectivism. Power distance did not influence the relations.
-
Gianfreda, Angelica; Ravazzolo, Francesco & Rossini, Luca
(2022)
Large Time-Varying Volatility Models for Hourly Electricity Prices*
Vis sammendrag
We study the importance of time-varying volatility in modelling hourly electricity prices when fundamental drivers are included in the estimation. This allows us to contribute to the literature of large Bayesian VARs by using well-known time series models in a large dimension for the matrix of coefficients. Based on novel Bayesian techniques, we exploit the importance of both Gaussian and non-Gaussian error terms in stochastic volatility. We find that using regressors as fuel prices, forecasted demand and forecasted renewable energy is essential to properly capture the volatility of these prices. Moreover, we show that the time-varying volatility models outperform the constant volatility models in both the in-sample model-fit and the out-of-sample forecasting performance.
-
Huber, Andreas; Schröder, Daniel Thilo, Pogorelov, Konstantin, Griwodz, Carsten & Langguth, Johannes
(2022)
A Streaming System for Large-scale Temporal Graph Mining of Reddit Data
-
Ivanovska, Magdalena & Slavkovik, Marija
(2022)
Probabilistic Judgement Aggregation by Opinion Update
Vis sammendrag
We consider a situation where agents are updating their probabilistic opinions on a set of issues with respect to the confidence they have in each other’s judgements. We adapt the framework for reaching a consensus introduced in [2] and modified in [1] to our case of uncertain probabilistic judgements on logically related issues. We discuss possible alternative solutions for the instances where the requirements for reaching a consensus are not satisfied.
-
Moss, Jonas
(2022)
Infinite diameter confidence sets in Hedges’ publication bias model
Vis sammendrag
Meta-analysis, the statistical analysis of results from separate studies, is a fundamental building block of science. But the assumptions of classical meta-analysis models are not satisfied whenever publication bias is present, which causes inconsistent parameter estimates. Hedges’ selection function model takes publication bias into account, but estimating and inferring with this model is tough for some datasets. Using a generalized Gleser–Hwang theorem, we show there is no confidence set of guaranteed finite diameter for the parameters of Hedges’ selection model. This result provides a partial explanation for why inference with Hedges’ selection model is fraught with difficulties.
-
Yang, Wei-Ting; Reis, Marco, Borodin, Valeria, Juge, Michel & Roussy, Agnès
(2022)
An interpretable unsupervised Bayesian network model for fault detection and diagnosis
Vis sammendrag
Process monitoring is a critical activity in manufacturing industries. A wide variety of data-driven approaches have been developed and employed for fault detection and fault diagnosis. Analyzing the existing process monitoring schemes, prediction accuracy of the process status is usually the primary focus while the explanation (diagnosis) of a detected fault is relegated to a secondary role. In this paper, an interpretable unsupervised machine learning model based on Bayesian Networks (BN) is proposed to be the fundamental model supporting the process monitoring scheme. The proposed methodology is aligned with the recent efforts of eXplanatory Artificial Intelligence (XAI) for knowledge induction and decision making, now brought to the scope of advanced process monitoring. A BN is capable of combining data-driven induction with existing domain knowledge about the process and to display the underlying causal interactions of a process system in an easily interpretable graphical form. The proposed fault detection scheme consists of two levels of monitoring. In the first level, a global index is computed and monitored to detect any deviation from normal operation conditions. In the second level, two local indices are proposed to examine the fine structure of the fault, once it is signaled at the first level. These local indices support the diagnosis of the fault, and are based on the individual unconditional and conditional distributions of the monitored variables. A new labeling procedure is also proposed to narrow down the search and identify the fault type. Unlike many existing diagnosis methods that require access to faulty data (supervised diagnosis methods), the proposed diagnosis methodology belongs to the class that only requires data under normal conditions (unsupervised diagnosis methods). The effectiveness of the proposed monitoring scheme is demonstrated and validated through simulated datasets and an industrial dataset from semiconductor manufacturing.
-
Lundén, Daniel; Öhman, Joey, Kudlicka, Jan, Senderov, Viktor, Ronquist, Fredrik & Broman, David
(2022)
Compiling Universal Probabilistic Programming Languages with Efficient Parallel Sequential Monte Carlo Inference
Vis sammendrag
Probabilistic programming languages (PPLs) allow users to encode arbitrary inference problems, and PPL implementations provide general-purpose automatic inference for these problems. However, constructing inference implementations that are efficient enough is challenging for many real-world problems. Often, this is due to PPLs not fully exploiting available parallelization and optimization opportunities. For example, handling probabilistic checkpoints in PPLs through continuation-passing style transformations or non-preemptive multitasking—as is done in many popular PPLs—often disallows compilation to low-level languages required for high-performance platforms such as GPUs. To solve the checkpoint problem, we introduce the concept of PPL control-flow graphs (PCFGs)—a simple and efficient approach to checkpoints in low-level languages. We use this approach to implement RootPPL: a low-level PPL built on CUDA and C++ with OpenMP, providing highly efficient and massively parallel SMC inference. We also introduce a general method of compiling universal high-level PPLs to PCFGs and illustrate its application when compiling Miking CorePPL—a high-level universal PPL—to RootPPL. The approach is the first to compile a universal PPL to GPUs with SMC inference. We evaluate RootPPL and the CorePPL compiler through a set of real-world experiments in the domains of phylogenetics and epidemiology, demonstrating up to 6× speedups over state-of-the-art PPLs implementing SMC inference.
-
Durante, Fabrizio; Gianfreda, Angelica, Ravazzolo, Francesco & Rossini, Luca
(2022)
A multivariate dependence analysis for electricity prices, demand and renewable energy sources
Vis sammendrag
This paper examines the dependence between electricity prices, demand, and renewable energy sources by means of a multivariate copula model while studying Germany, the widest studied market in Europe. The inter-dependencies are investigated in-depth and monitored over time, with particular emphasis on the tail behavior. To this end, suitable tail dependence measures are introduced to take into account a multivariate extreme scenario appropriately identified through the Kendall’s distribution function. The empirical evidence demonstrates a strong association between electricity prices, renewable energy sources, and demand within a day and over the studied years. Hence, this analysis provides guidance for further and different incentives for promoting green energy generation while considering the time-varying dependencies of the involved variables.
-
Billé, Anna Gloria; Gianfreda, Angelica, Grosso, Filippo Del & Ravazzolo, Francesco
(2022)
Forecasting electricity prices with expert, linear, and nonlinear models
Vis sammendrag
This paper compares several models for forecasting regional hourly day-ahead electricity prices, while accounting for fundamental drivers. Forecasts of demand, in-feed from renewable energy sources, fossil fuel prices, and physical flows are all included in linear and nonlinear specifications, ranging in the class of ARFIMA-GARCH models—hence including parsimonious autoregressive specifications (known as expert-type models). The results support the adoption of a simple structure that is able to adapt to market conditions. Indeed, we include forecasted demand, wind and solar power, actual generation from hydro, biomass, and waste, weighted imports, and traditional fossil fuels. The inclusion of these exogenous regressors, in both the conditional mean and variance equations, outperforms in point and, especially, in density forecasting when the superior set of models is considered. Indeed, using the model confidence set and considering northern Italian prices, predictions indicate the strong predictive power of regressors, in particular in an expert model augmented for GARCH-type time-varying volatility. Finally, we find that using professional and more timely predictions of consumption and renewable energy sources improves the forecast accuracy of electricity prices more than using predictions publicly available to researchers.
-
Avesani, Diego; Zanfei, Ariele, Marco, Nicola Di, Galletti, Andrea, Ravazzolo, Francesco, Righetti, Maurizio & Majone, Bruno
(2022)
Short-term hydropower optimization driven by innovative time-adapting econometric model
Vis sammendrag
The ongoing transformation of the electricity market has reshaped the hydropower production paradigm for storage reservoir systems, with a shift from strategies oriented towards maximizing regional energy production to strategies aimed at the revenue maximization of individual systems. Indeed, hydropower producers bid their energy production scheduling 1 day in advance, attempting to align the operational plan with hours where the expected electricity prices are higher. As a result, the accuracy of 1-day ahead prices forecasts has started to play a key role in the short-term optimization of storage reservoir systems. This paper aims to contribute to the topic by presenting a comparative assessment of revenues provided by short-term optimizations driven by two econometric models. Both models are autoregressive time-adapting hourly forecasting models, which exploit the information provided by past values of electricity prices, with one model, referred to as Autoarimax, additionally considering exogenous variables related to electricity demand and production. The benefit of using the innovative Autoarimax model is exemplified in two selected hydropower systems with different storage capacities. The enhanced accuracy of electricity prices forecasting is not constant across the year due to the large uncertainties characterizing the electricity market. Our results also show that the adoption of Autoarimax leads to larger revenues with respect to the use of a standard model, increases that depend strongly on the hydropower system characteristics. Our results may be beneficial for hydropower companies to enhance the expected revenues from storage hydropower systems, especially those characterized by large storage capacity.
-
Pogorelov, Konstantin; Schroeder, Daniel Thilo, Brenner, Stefan, Maulana, Asep & Langguth, Johannes
(2022)
Combining Tweets and Connections Graph for FakeNews Detection at MediaEval 2022
CEUR Workshop Proceedings.
-
Caporin, Massimiliano; Gupta, Rangan & Ravazzolo, Francesco
(2021)
Contagion between real estate and financial markets: A Bayesian quantile-on-quantile approach
Vis sammendrag
We study contagion between Real Estate Investment Trusts (REITs) and the equity market in the U.S. over four sub-samples covering January, 2003 to December, 2017, by using Bayesian nonparametric quantile-on-quantile (QQ) regressions with heteroskedasticity. We find that the spillovers from the REITs on to the equity market has varied over time and quantiles defining the states of these two markets across the four sub-samples, thus providing evidence of shift-contagion. Further, contagion from REITs upon the stock market went up during the global financial crisis particularly, and also over the period corresponding to the European sovereign debt crisis, relative to the pre-crisis period. Our main findings are robust to alternative model specifications of the benchmark Bayesian QQ model, especially when we control for omitted variable bias using the heteroskedastic error structure. Our results have important implications for various agents in the economy namely, academics, investors and policymakers.
-
Pogorelov, Konstantin; Schroeder, Daniel Thilo, Brenner, Stefan & Langguth, Johannes
(2021)
FakeNews: Corona Virus and Conspiracies Multimedia Analysis Task at MediaEval 2021
CEUR Workshop Proceedings.
-
Ferrari, Davide; Ravazzolo, Francesco & Vespignani, Joaquin
(2021)
Forecasting energy commodity prices: A large global dataset sparse approach
Vis sammendrag
This paper focuses on forecasting quarterly nominal global energy prices of commodities, such as oil, gas and coal,using the Global VAR dataset proposed by Mohaddes and Raissi (2018). This dataset includes a number of poten-tially informative quarterly macroeconomic variables for the 33 largest economies, overall accounting for morethan 80% of the global GDP. To deal with the information on this large database, we apply dynamic factor modelsbased on a penalized maximum likelihood approach that allows to shrink parameters to zero and to estimatesparse factor loadings. The estimated latent factors show considerable sparsity and heterogeneity in the selectedloadings across variables. When the model is extended to predict energy commodity prices up to four periodsahead, results indicate larger predictability relative to the benchmark random walk model for 1-quarter aheadfor all energy commodities and up to 4 quarters ahead for gas prices. Our model also provides superior forecaststhan machine learning techniques, such as elastic net, LASSO and random forest, applied to the same database.
-
Yazidi, Anis; Ivanovska, Magdalena, Zennaro, Fabio Massimo, Lind, Pedro & Viedma, Enrique Herrera
(2021)
A new decision making model based on Rank Centrality for GDM with fuzzy preference relations
Vis sammendrag
Preference aggregation in Group Decision Making (GDM) is a substantial problem that has received a lot of research attention. Decision problems involving fuzzy preference relations constitute an important class within GDM. Legacy approaches dealing with the latter type of problems can be classified into indirect approaches, which involve deriving a group preference matrix as an intermediate step, and direct approaches, which deduce a group preference ranking based on individual preference rankings. Although the work on indirect approaches has been extensive in the literature, there is still a scarcity of research dealing with the direct approaches. In this paper we present a direct approach towards aggregating several fuzzy preference relations on a set of alternatives into a single weighted ranking of the alternatives. By mapping the pairwise preferences into transitions probabilities, we are able to derive a preference ranking from the stationary distribution of a stochastic matrix. Interestingly, the ranking of the alternatives obtained with our method corresponds to the optimizer of the Maximum Likelihood Estimation of a particular Bradley-Terry-Luce model. Furthermore, we perform a theoretical sensitivity analysis of the proposed method supported by experimental results and illustrate our approach towards GDM with a concrete numerical example. This work opens avenues for solving GDM problems using elements of probability theory, and thus, provides a sound theoretical fundament as well as plausible statistical interpretation for the aggregation of expert opinions in GDM.
-
Schroeder, Daniel Thilo; Lind, Pedro, Pogorelov, Konstantin & Langguth, Johannes
(2021)
A Framework for Interaction-based Propagation Analysis in Online Social Networks
-
Burchard, Luk; Moe, Johannes Sellæg, Schroeder, Daniel Thilo, Pogorelov, Konstantin & Langguth, Johannes
(2021)
iPUG: Accelerating Breadth-First Graph Traversals Using Manycore Graphcore IPUs
Vis sammendrag
The Graphcore Intelligence Processing Unit (IPU) is a newly developed processor type whose architecture does not rely on the traditional caching hierarchies. Developed to meet the need for more and more data-centric applications, such as machine learning, IPUs combine a dedicated portion of SRAM with each of its numerous cores, resulting in high memory bandwidth at the price of capacity. The proximity of processor cores and memory makes the IPU a promising field of experimentation for graph algorithms since it is the unpredictable, irregular memory accesses that lead to performance losses in traditional processors with pre-caching.
This paper aims to test the IPU’s suitability for algorithms with hard-to-predict memory accesses by implementing a breadth-first search (BFS) that complies with the Graph500 specifications. Precisely because of its apparent simplicity, BFS is an established benchmark that is not only subroutine for a variety of more complex graph algorithms, but also allows comparability across a wide range of architectures.
We benchmark our IPU code on a wide range of instances and compare its performance to state-of-the-art CPU and GPU codes. The results indicate that the IPU delivers speedups of up to 4× over the fastest competing result on an NVIDIA V100 GPU, with typical speedups of about 1.5× on most test instances.
-
Burchard, Luk; Cai, Xing & Langguth, Johannes
(2021)
iPUG for Multiple Graphcore IPUs: Optimizing Performance and Scalability of Parallel Breadth-First Search
-
Hjort, Nils Lid & Stoltenberg, Emil Aas
(2021)
The partly parametric and partly nonparametric additive risk model
Vis sammendrag
Aalen’s linear hazard rate regression model is a useful and increasingly popular alternative to Cox’ multiplicative hazard rate model. It postulates that an individual has hazard rate function h(s)=z1α1(s)+⋯+zrαr(s) in terms of his covariate values z1,…,zr. These are typically levels of various hazard factors, and may also be time-dependent. The hazard factor functions αj(s) are the parameters of the model and are estimated from data. This is traditionally accomplished in a fully nonparametric way. This paper develops methodology for estimating the hazard factor functions when some of them are modelled parametrically while the others are left unspecified. Large-sample results are reached inside this partly parametric, partly nonparametric framework, which also enables us to assess the goodness of fit of the model’s parametric components. In addition, these results are used to pinpoint how much precision is gained, using the parametric-nonparametric model, over the standard nonparametric method. A real-data application is included, along with a brief simulation study.
-
Schroeder, Daniel Thilo; Schaal, Ferdinand, Filkukova, Petra, Pogorelov, Konstantin & Langguth, Johannes
(2021)
WICO Graph: A Labeled Dataset of Twitter Subgraphs based on Conspiracy Theory and 5G-Corona Misinformation Tweets.
Proceedings of the International Conference on Agents and Artificial Intelligence (ICAART).
Doi:
10.5220/0010262802570266
-
Agudze, Komla M.; Billio, Monica, Casarin, Roberto & Ravazzolo, Francesco
(2021)
Markov switching panel with endogenous synchronization effects
Vis sammendrag
This paper introduces a new dynamic panel model with multi-layer network effects. Series-specific latent Markov chain processes drive the dynamics of the observable processes, and several types of interaction effects among the hidden chains allow for various degrees of endogenous synchronization of both latent and observable processes. The interaction is driven by a multi-layer network with exogenous and endogenous connectivity layers. We provide some theoretical properties of the model, develop a Bayesian inference framework and an efficient Markov Chain Monte Carlo algorithm for estimating parameters, latent states, and endogenous network layers. An application to the US-state coincident indicators shows that the synchronization in the US economy is generated by network effects among the states. The inclusion of a multi-layer network provides a new tool for measuring the effects of the public policies that impact the connectivity between the US states, such as mobility restrictions or job support schemes. The proposed new model and the related inference are general and may find application in a wide spectrum of datasets where the extraction of endogenous interaction effects is relevant and of interest.
-
Burchard, Luk Bjarne; Moe, Johannes, Schroeder, Daniel Thilo, Pogorelov, Konstantin & Langguth, Johannes
(2021)
iPUG: Accelerating Breadth-First Graph Traversals Using Manycore Graphcore IPUs
-
Ravazzolo, Francesco & Vespignani, Joaquin
(2020)
World steel production: A new monthly indicator of global real economic activity
-
Pogorelov, Konstantin; Schroeder, Daniel Thilo, Burchard, Luk Bjarne, Moe, Johannes, Brenner, Stefan, Filkukova, Petra & Langguth, Johannes
(2020)
FakeNews: Corona Virus and 5G Conspiracy Task at MediaEval 2020
CEUR Workshop Proceedings.
-
Pogorelov, Konstantin; Schroeder, Daniel Thilo, Filkukova, Petra & Langguth, Johannes
(2020)
A System for High Performance Mining on GDELT Data
-
Schroeder, Daniel Thilo; Pogorelov, Konstantin & Langguth, Johannes
(2020)
Evaluating Standard Classifiers for Detecting COVID-19 related Misinformation
CEUR Workshop Proceedings.
-
Concetto, Chiara Limongi & Ravazzolo, Francesco
(2019)
Optimism in Financial Markets: Stock Market Returns and Investor Sentiments
-
Vassøy, Bjørnar; Ruocco, Massimiliano, Silva, Eliezer de Souza da & Aune, Erlend
(2019)
Time is of the essence: A joint Hierarchical RNN and Point Process model for time and item predictions
-
Schroeder, Daniel Thilo; Pogorelov, Konstantin & Langguth, Johannes
(2019)
Fact: a framework for analysis and capture of twitter graphs
-
Bassetti, Federico; Casarin, Roberto & Ravazzolo, Francesco
(2018)
Bayesian Nonparametric Calibration and Combination of Predictive Distributions
-
Yang, Wei-Ting; Blue, Jakey, Roussy, Agnès, Reis, Marco & Pinaton, Jacques
(2018)
Virtual metrology modeling based on gaussian bayesian network
-
Clark, Todd E. & Ravazzolo, Francesco
(2015)
Macroeconomic Forecasting Performance under Alternative Specifications of Time-Varying Volatility