Vincent François-Lavet

An energy-based variational model of ferromagnetic hysteresis for finite element computations

This paper proposes a macroscopic model for ferromagnetic hysteresis that is well-suited for finite element implementation. The model is readily vectorial and relies on a consistent thermodynamic formulation. In particular, the stored magnetic energy and the dissipated energy are known at all times, and not solely after the completion of closed hysteresis loops as is usually the case. The obtained incremental formulation is variationally consistent, i.e., all internal variables follow from the minimization of a thermodynamic potential.

Simple connectome inference from partial correlation statistics in calcium imaging

This book illustrates the thrust of the scientific community to use machine learning concepts for tackling a complex problem: given time series of neuronal spontaneous activity, which is the underlying connectivity between the neurons in the network? The contributing authors also develop tools for the advancement of neuroscience through machine learning techniques, with a focus on the major open problems in neuroscience. While the techniques have been developed for a specific application, they address the more general problem of network reconstruction from observational time series, a problem of interest in a wide variety of domains, including econometrics, epidemiology, and climatology, to cite only a few. The book is designed for the mathematics, physics and computer science communities that carry out research in neuroscience problems. The content is also suitable for the machine learning community because it exemplifies how to approach the same problem from different perspectives.

Imitative Learning for Online Planning in Microgrids

This paper aims to design an algorithm dedicated to operational planning for microgrids in the challenging case where the scenarios of production and consumption are not known in advance. Using expert knowledge obtained from solving a family of linear programs, we build a learning set for training a decision-making agent. The empirical performances in terms of Levelized Energy Cost LEC of the obtained agent are compared to the expert performances obtained in the case where the scenarios are known in advance. Preliminary results are promising.

Using approximate dynamic programming for estimating the revenues of a hydrogen-based high-capacity storage device

This paper proposes a methodology to estimate the maximum revenue that can be generated by a company that operates a high-capacity storage device to buy or sell electricity on the day-ahead electricity market. The methodology exploits the Dynamic Programming (DP) principle and is specified for hydrogen-based storage devices that use electrolysis to produce hydrogen and fuel cells to generate electricity from hydrogen. Experimental results are generated using historical data of energy prices on the Belgian market. They show how the storage capacity and other parameters of the storage device influence the optimal revenue. The main conclusion drawn from the experiments is that it may be advisable to invest in large storage tanks to exploit the inter-seasonal price fluctuations of electricity.

Approximate Bayes Optimal Policy Search using Neural Networks

Bayesian Reinforcement Learning (BRL) agents aim to maximise the expected collected rewards obtained when interacting with an unknown Markov Decision Process (MDP) while using some prior knowledge. State-of-the-art BRL agents rely on frequent updates of the belief on the MDP, as new observations of the environment are made. This offers theoretical guarantees to converge to an optimum, but is computationally intractable, even on small-scale problems. In this paper, we present a method that circumvents this issue by training a parametric policy able to recommend an action directly from raw observations. Artificial Neural Networks (ANNs) are used to represent this policy, and are trained on the trajectories sampled from the prior. The trained model is then used online, and is able to act on the real MDP at a very low computational cost. Our new algorithm shows strong empirical performance, on a wide range of test problems, and is robust to inaccuracies of the prior distribution.