University of LiègeULgFaculty of EngineeringFacSALibrary News   
Gilles Louppe - Publications ORBI
Louppe, G., Hermans, J., & Cranmer, K. (2019). Adversarial Variational Optimization of Non-Differentiable Simulators. Proceedings of Machine Learning Research.
Peer reviewed
Complex computer simulators are increasingly used across fields of science as generative models tying parameters of an underlying theory to experimental observations. Inference in this setup is often difficult ...
Hermans, J., Begy, V., & Louppe, G. (2019). Likelihood-free MCMC with Approximate Likelihood Ratios. Eprint/Working paper retrieved from https://arxiv.org/abs/1903.04057.
We propose a novel approach for posterior sampling with intractable likelihoods. This is an increasingly important problem in scientific applications where models are implemented as sophisticated computer ...
Louppe, G., Cho, K., Becot, C., & Cranmer, K. (2019, January 07). QCD-Aware Recursive Neural Networks for Jet Physics. Journal of High Energy Physics.
Peer reviewed (verified by ORBi)
Recent progress in applying machine learning for jet physics has been built upon an analogy between calorimeters and images. In this work, we present a novel class of recursive neural networks built instead ...
Pesah, A., Wehenkel, A., & Louppe, G. (2018, December 08). Recurrent machines for likelihood-free inference. Paper presented at Workshop of Meta-Learning at Thirty-second Conference on Neural Information Processing Systems 2018, Montreal, Canada.
Peer reviewed
Likelihood-free inference is concerned with the estimation of the parameters of a non-differentiable stochastic simulator that best reproduce real observations. In the absence of a likelihood function, most ...
Sabatelli, M., Louppe, G., Geurts, P., & Wiering, M. (2018, December 07). Deep Quality Value (DQV) Learning. Advances in Neural Information Processing Systems, Deep Reinforcement Learning Workshop.
Peer reviewed
We introduce a novel Deep Reinforcement Learning (DRL) algorithm called Deep Quality-Value (DQV) Learning. DQV uses temporal-difference learning to train a Value neural network and uses this network for ...
Louppe, G. (2018, October 17). Constraining Effective Field Theories with Machine Learning. Paper presented at 3rd ATLAS Machine Learning Workshop, Geneva, Switzerland.
We present powerful new analysis techniques to constrain effective field theories at the LHC. By leveraging the structure of particle physics processes, we extract extra information from Monte-Carlo ...
Brehmer, J., Cranmer, K., Louppe, G., & Pavez, J. (2018, September 12). Constraining Effective Field Theories with Machine Learning. Physical Review Letters.
Peer reviewed (verified by ORBi)
We present powerful new analysis techniques to constrain effective field theories at the LHC. By leveraging the structure of particle physics processes, we extract extra information from Monte-Carlo ...
Brehmer, J., Cranmer, K., Louppe, G., & Pavez, J. (2018, September 12). A Guide to Constraining Effective Field Theories with Machine Learning. Physical Review. D.
Peer reviewed (verified by ORBi)
We develop, discuss, and compare several inference techniques to constrain theory parameters in collider experiments. By harnessing the latent-space structure of particle physics processes, we extract extra ...
Stoye, M., Brehmer, J., Louppe, G., Pavez, J., & Cranmer, K. (2018). Likelihood-free inference with an improved cross-entropy estimator. Eprint/Working paper retrieved from https://arxiv.org/abs/1808.00973.
We extend recent work (Brehmer, et. al., 2018) that use neural networks as surrogate models for likelihood-free inference. As in the previous work, we exploit the fact that the joint likelihood ratio and joint ...
Gunes Baydin, A., Heinrich, L., Bhimji, W., Gram-Hansen, B., Louppe, G., Shao, L., Prabhat, Cranmer, K., & Wood, F. (2018). Efficient Probabilistic Inference in the Quest for Physics Beyond the Standard Model. Eprint/Working paper retrieved from https://arxiv.org/abs/1807.07706.
We present a novel framework that enables efficient probabilistic inference in large-scale scientific models by allowing the execution of existing domain-specific simulators as probabilistic programs ...
Albertsson, K., Altoe, P., Anderson, D., Andrews, M., Araque Espinosa, J. P., Aurisano, A., Basara, L., Bevan, A., Bhimji, W., Bonacorsi, D., Calafiura, P., Campanelli, M., Capps, L., Carminati, F., Carrazza, S., Childers, T., Coniavitis, E., Cranmer, K., David, C., Davis, D., Duarte, J., Erdmann, M., Eschle, J., Farbin, A., Feickert, M., Filipe Castro, N., Fitzpatrick, C., Floris, M., Forti, A., Garra-Tico, J., Gemmler, J., Girone, M., Glaysher, P., Gleyzer, S., Gligorov, V., Golling, T., Graw, J., Gray, L., Greenwood, D., Hacker, T., Harvey, J., Hegner, B., Heinrich, L., Hooberman, B., Junggeburth, J., Kagan, M., Kane, M., Kanishchev, K., Karpiński, P., Kassabov, Z., Kaul, G., Kcira, D., Keck, T., Klimentov, A., Kowalkowski, J., Kreczko, L., Kurepin, A., Kutschke, R., Kuznetsov, V., Köhler, N., Lakomov, I., Lannon, K., Lassnig, M., Limosani, A., Louppe, G., Mangu, A., Mato, P., Meenakshi, N., Meinhard, H., Menasce, D., Moneta, L., Moortgat, S., Neubauer, M., Newman, H., Pabst, H., Paganini, M., Paulini, M., Perdue, G., Perez, U., Picazio, A., Pivarski, J., Prosper, H., Psihas, F., Radovic, A., Reece, R., Rinkevicius, A., Rodrigues, E., Rorie, J., Rousseau, D., Sauers, A., Schramm, S., Schwartzman, A., Severini, H., Seyfert, P., Siroky, F., Skazytkin, K., Sokoloff, M., Stewart, G., Stienen, B., Stockdale, I., Strong, G., Thais, S., Tomko, K., Upfal, E., Usai, E., Ustyuzhanin, A., Vala, M., Vallecorsa, S., Verzetti, M., Vilasís-Cardona, X., Vlimant, J.-R., Vukotic, I., Wang, S.-J., Watts, G., Williams, M., Wu, W., Wunsch, S., & Zapata, O. (2018). Machine Learning in High Energy Physics Community White Paper. Journal of Physics. Conference Series, 1085.
Peer reviewed (verified by ORBi)
Machine learning is an important research area in particle physics, beginning with applications to high-level physics analysis in the 1990s and 2000s, followed by an explosion of applications in particle and ...
Hermans, J., & Louppe, G. (2018). Gradient Energy Matching for Distributed Asynchronous Gradient Descent. Eprint/Working paper retrieved from https://arxiv.org/abs/1805.08469.
Distributed asynchronous SGD has become widely used for deep learning in large-scale systems, but remains notorious for its instability when increasing the number of workers. In this work, we study the ...
Brehmer, J., Louppe, G., Pavez, J., & Cranmer, K. (2018). Mining gold from implicit models to improve likelihood-free inference. Eprint/Working paper retrieved from https://arxiv.org/abs/1805.12244.
Simulators often provide the best description of real-world phenomena; however, they also lead to challenging inverse problems because the density they implicitly define is often intractable. We present a new ...
Engemann, D. A., Raimondo, F., King, J.-R., Rohaut, B., Louppe, G., Faugeras, F., Annen, J., Cassol, H., Gosseries, O., Fernandez-Slezak, D., Laureys, S., Naccache, L., Dehaene, S., & Sitt, J. D. (2018). Robust EEG-based cross-site and cross-protocol classification of states of consciousness. Brain: a Journal of Neurology, 141(11), 3179-3192.
Peer reviewed (verified by ORBi)
Determining the state of consciousness in patients with disorders of consciousness is a challenging practical and theoretical problem. Recent findings suggest that multiple markers of brain activity extracted ...
Cranmer, K., Pavez, J., Louppe, G., & Brooks, W. K. (2016). Experiments using machine learning to approximate likelihood ratios for mixture models. Journal of Physics Conference Series.
Peer reviewed
Likelihood ratio tests are a key tool in many fields of science. In order to evaluate the likelihood ratio the likelihood function is needed. However, it is common in fields such as High Energy Physics to have ...
Louppe, G., Kagan, M., & Cranmer, K. (2016). Learning to Pivot with Adversarial Networks. Advances in Neural Information Processing Systems.
Peer reviewed
Several techniques for domain adaptation have been proposed to account for differences in the distribution of the data used for training and testing. The majority of this work focuses on a binary domain label ...
Maguire, E., Montull, J. M., & Louppe, G. (2016). Visualization of Publication Impact. EuroVis '16 Proceedings of the Eurographics / IEEE VGTC Conference on Visualization: Short Papers.
Peer reviewed
Measuring scholarly impact has been a topic of much interest in recent years. While many use the citation count as a primary indicator of a publications impact, the quality and impact of those citations will ...
Marée, R., Rollus, L., Stévens, B., Hoyoux, R., Louppe, G., Vandaele, R., Begon, J.-M., Kainz, P., Geurts, P., & Wehenkel, L. (2016, January 10). Collaborative analysis of multi-gigapixel imaging data using Cytomine. Bioinformatics, 7.
Peer reviewed (verified by ORBi)
Motivation: Collaborative analysis of massive imaging datasets is essential to enable scientific discoveries. Results: We developed Cytomine to foster active and distributed collaboration of ...
Louppe, G., Al-Natsheh, H., Susik, M., & Maguire, E. (2015). Ethnicity sensitive author disambiguation using semi-supervised learning. Communications in Computer and Information Science.
Peer reviewed
Author name disambiguation in bibliographic databases is the problem of grouping together scientific publications written by the same person, accounting for potential homonyms and/or synonyms. Among solutions ...
Cranmer, K., Pavez, J., & Louppe, G. (2015). Approximating Likelihood Ratios with Calibrated Discriminative Classifiers. Eprint/Working paper retrieved from https://arxiv.org/abs/1506.02169.
In many fields of science, generalized likelihood ratio tests are established tools for statistical inference. At the same time, it has become increasingly common that a simulator (or generative model) is used ...
Louppe, G. (2015, April 03). Tree models with Scikit-Learn: Great models with little assumptions. Paper presented at PyData Paris 2015, Paris, France.
This talk gives an introduction to tree-based methods, both from a theoretical and practical point of view. It covers decision trees, random forests and boosting estimators, along with concrete examples based ...
McGovern, A., Gagne II, D. J., Eustaquio, L., Titericz, G., Lazorthes, B., Zhang, O., Louppe, G., Prettenhofer, P., Basara, J., Hamill, T. M., & Margolin, D. (2015). Solar Energy Prediction: An International Contest to Initiate Interdisciplinary Research on Compelling Meteorological Problems. Bulletin of the American Meteorological Society.
Peer reviewed (verified by ORBi)
As meteorological observing systems and models grow in complexity and number, the size of the data becomes overwhelming for humans to analyze using traditional techniques. Computer scientists, and specifically ...
Louppe, G. (2014). Understanding Random Forests: From Theory to Practice. Unpublished doctoral thesis, Université de Liège, ​Liège, ​​Belgique.
Data analysis and machine learning have become an integrative part of the modern scientific methodology, offering automated procedures for the prediction of a phenomenon based on past observations, unraveling ...
Louppe, G. (2014, August 29). Accelerating Random Forests in Scikit-Learn. Paper presented at EuroScipy 2014, Cambridge, UK.
Peer reviewed
Random Forests are without contest one of the most robust, accurate and versatile tools for solving machine learning tasks. Implementing this algorithm properly and efficiently remains however a challenging ...
Sutera, A., Joly, A., François-Lavet, V., Qiu, Z., Louppe, G., Ernst, D., & Geurts, P. (2014). Simple connectome inference from partial correlation statistics in calcium imaging. In J., Soriano, D., Battaglia, I., Guyon, V., Lemaire, J., Orlandi, & B., Ray (Eds.), Neural Connectomics Challenge. Springer.
Peer reviewed
In this work, we propose a simple yet effective solution to the problem of connectome inference in calcium imaging data. The proposed algorithm consists of two steps. First, processing the raw signals to ...
Marée, R., Rollus, L., Stevens, B., Louppe, G., Caubo, O., Rocks, N., Bekaert, S., Cataldo, D., & Wehenkel, L. (2014). A hybrid human-computer approach for large-scale image-based measurements using web services and machine learning. Proceedings IEEE International Symposium on Biomedical Imaging. IEEE.
Peer reviewed
We present a novel methodology combining web-based software development practices, machine learning, and spatial databases for computer-aided quantification of regions of interest (ROIs) in large-scale ...
Botta, V., Louppe, G., Geurts, P., & Wehenkel, L. (2014, April 02). Exploiting SNP Correlations within Random Forest for Genome-Wide Association Studies. PLoS ONE.
Peer reviewed (verified by ORBi)
The primary goal of genome-wide association studies (GWAS) is to discover variants that could lead, in isolation or in combination, to a particular trait or disease. Standard approaches to GWAS, however, are ...
Prettenhofer, P., & Louppe, G. (2014, February 23). Gradient Boosted Regression Trees in Scikit-Learn. Paper presented at PyData 2014, London, UK.
This talk describes Gradient Boosted Regression Trees (GBRT), a powerful statistical learning technique with applications in a variety of areas, ranging from web page ranking to environmental niche modeling ...
Louppe, G., & Prettenhofer, P. (2014, February 05). Forecasting Daily Solar Energy Production Using Robust Regression Techniques. Paper presented at 94th American Meteorological Society Annual Meeting, Atlanta, USA.
We describe a novel approach to forecast daily solar energy production based on the output of a numerical weather prediction (NWP) model using non-parametric robust regression techniques. Our approach ...
Joly, A., & Louppe, G. (2014, January 27). Scikit-Learn: Machine Learning in the Python ecosystem. Poster session presented at GIGA DAY 2014, Liège, Belgium.
The scikit-learn project is an increasingly popular machine learning library written in Python. It is designed to be simple and efficient, useful to both experts and non-experts, and reusable in a variety of ...
Louppe, G., & Varoquaux, G. (2013, December 10). Scikit-Learn: Machine Learning in the Python ecosystem. Paper presented at NIPS 2013 Workshop on Machine Learning Open Source Software.
Peer reviewed
The scikit-learn project is an increasingly popular machine learning library written in Python. It is designed to be simple and efficient, useful to both experts and non-experts, and reusable in a variety of ...
Louppe, G., Wehenkel, L., Sutera, A., & Geurts, P. (2013). Understanding variable importances in forests of randomized trees. Advances in Neural Information Processing Systems 26.
Peer reviewed
Despite growing interest and practical use in various scientific areas, variable importances derived from tree-based ensemble methods are not well understood from a theoretical point of view. In this work we ...
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., & Duchesnay, É. (2012). Scikit-learn: Machine Learning in Python. arXiv e-prints, 1201.
Peer reviewed
Scikit-learn is a Python module integrating a wide range of state-of-the-art machine learning algorithms for medium-scale supervised and unsupervised problems. This package focuses on bringing machine learning ...
Louppe, G., & Geurts, P. (2012). Ensembles on Random Patches. Machine Learning and Knowledge Discovery in Databases. Berlin, Germany: Springer-Verlag.
Peer reviewed
In this paper, we consider supervised learning under the assumption that the available memory is small compared to the dataset size. This general framework is relevant in the context of big data, distributed ...
Geurts, P., & Louppe, G. (2011). Learning to rank with extremely randomized trees. Proceedings of Machine Learning Research, 14, 49-61.
Peer reviewed
In this paper, we report on our experiments on the Yahoo! Labs Learning to Rank challenge organized in the context of the 23rd International Conference of Machine Learning (ICML 2010). We competed in both the ...
Louppe, G., & Geurts, P. (2010, December 11). A zealous parallel gradient descent algorithm. Poster session presented at NIPS 2010 Workshop on Learning on Cores, Clusters and Clouds, Whistler, Canada.
Peer reviewed
Parallel and distributed algorithms have become a necessity in modern machine learning tasks. In this work, we focus on parallel asynchronous gradient descent and propose a zealous variant that minimizes the ...
Louppe, G. (2010). Collaborative filtering: Scalable approaches using restricted Boltzmann machines. Unpublished master thesis, Université de Liège, ​Liège, ​​Belgique.
Parallel to the growth of electronic commerce, recommender systems have become a very active area of research, both in the industry and in the academic world. The goal of these systems is to make automatic but ...