5 papers
Neural delay differential equations: learning non-Markovian closures for partially known dynamical systems
Thibault Monsel, Onofrio Semeraro, Lionel Mathelin +1
Recent advances in learning dynamical systems from data have shown significant promise. However, many existing methods assume access to the full state of the system -- an assumptio…
Latent-Space Non-Linear Model Predictive Control for Partially-Observable Systems
Luigi Marra, Onofrio Semeraro, Lionel Mathelin +2
This work presents a scalable control framework based on nonlinear Model Predictive Control for high-dimensional dynamical systems. The proposed approach addresses the key challeng…
Learning non-Markovian Dynamical Systems with Signature-based Encoders
Eliott Pradeleix, Rémy Hosseinkhan-Boucher, Alena Shilova +2
Neural ordinary differential equations offer an effective framework for modeling dynamical systems by learning a continuous-time vector field. However, they rely on the Markovian a…
Increasing Information for Model Predictive Control with Semi-Markov Decision Processes
Rémy Hosseinkhan-Boucher, Onofrio Semeraro, Lionel Mathelin
Recent works in Learning-Based Model Predictive Control of dynamical systems show impressive sample complexity performances using criteria from Information Theory to accelerate the…
Evidence on the Regularisation Properties of Maximum-Entropy Reinforcement Learning
Rémy Hosseinkhan-Boucher, Onofrio Semeraro, Lionel Mathelin
The generalisation and robustness properties of policies learnt through Maximum-Entropy Reinforcement Learning are investigated on chaotic dynamical systems with Gaussian noise on…