Enhancing Reinforcement Learning Agents with Local Guides

1 code implementation • 21 Feb 2024 • Paul Daoudi, Bogdan Robu, Christophe Prieur, Ludovic Dos Santos, Merwan Barlier

This paper addresses the problem of integrating local guide policies into a Reinforcement Learning agent.

reinforcement-learning

Improving a Proportional Integral Controller with Reinforcement Learning on a Throttle Valve Benchmark

no code implementations • 21 Feb 2024 • Paul Daoudi, Bojan Mavkov, Bogdan Robu, Christophe Prieur, Emmanuel Witrant, Merwan Barlier, Ludovic Dos Santos

This paper presents a learning-based control strategy for non-linear throttle valves with an asymmetric hysteresis, leading to a near-optimal controller without requiring any prior knowledge about the environment.

Reinforcement Learning (RL)

A Conservative Approach for Few-Shot Transfer in Off-Dynamics Reinforcement Learning

no code implementations • 24 Dec 2023 • Paul Daoudi, Christophe Prieur, Bogdan Robu, Merwan Barlier, Ludovic Dos Santos

In the few-shot framework, a limited number of transitions from the target environment are introduced to facilitate a more effective transfer.

Imitation Learning

Conservative Exploration for Policy Optimization via Off-Policy Policy Evaluation

no code implementations • 24 Dec 2023 • Paul Daoudi, Mathias Formoso, Othman Gaizi, Achraf Azize, Evrard Garcelon

A precondition for the deployment of a Reinforcement Learning agent to a real-world system is to provide guarantees on the learning process.

Density Estimation for Conservative Q-Learning

no code implementations • 29 Sep 2021 • Paul Daoudi, Merwan Barlier, Ludovic Dos Santos, Aladin Virmaux

We hence introduce Density Conservative Q-Learning (D-CQL), a batch-RL algorithm with strong theoretical guarantees that carefully penalizes the value function based on the amount of information collected in the state-action space.

Density Estimation Q-Learning