Discovering Individual Rewards in Collective Behavior through Inverse Multi-Agent Reinforcement Learning

no code implementations • 17 May 2023 • Daniel Waelchli, Pascal Weber, Petros Koumoutsakos

In this study, we tackle this challenge by introducing an off-policy inverse multi-agent reinforcement learning algorithm (IMARL).

Multi-agent Reinforcement Learning OpenAI Gym +1

Deep Clustering With Consensus Representations

no code implementations • 13 Oct 2022 • Lukas Miklautz, Martin Teuffenbach, Pascal Weber, Rona Perjuci, Walid Durani, Christian Böhm, Claudia Plant

Further, we propose DECCS (Deep Embedded Clustering with Consensus representationS), a deep consensus clustering method that learns a consensus representation by enhancing the embedded space to such a degree that all ensemble members agree on a common clustering result.

Clustering Clustering Ensemble +1

Remember and Forget Experience Replay for Multi-Agent Reinforcement Learning

no code implementations • 24 Mar 2022 • Pascal Weber, Daniel Wälchli, Mustafa Zeqiri, Petros Koumoutsakos

We present the extension of the Remember and Forget for Experience Replay (ReF-ER) algorithm to Multi-Agent Reinforcement Learning (MARL).

Continuous Control Multi-agent Reinforcement Learning +3

Learning swimming escape patterns for larval fish under energy constraints

no code implementations • 3 May 2021 • Ioannis Mandralis, Pascal Weber, Guido Novati, Petros Koumoutsakos

The present, data efficient, reinforcement learning algorithm results in an array of patterns that reveal practical flow optimization principles for efficient swimming and the methodology can be transferred to the control of aquatic robotic devices operating under energy constraints.

reinforcement-learning Reinforcement Learning (RL) +1