5 papers
Bridging the Performance Gap Between Target-Free and Target-Based Reinforcement Learning
Théo Vincent, Yogesh Tripathi, Tim Faust +5
The use of target networks in deep reinforcement learning is a widely popular solution to mitigate the brittleness of semi-gradient approaches and stabilize learning. However, targ…
Learning what to say and how precisely: Efficient Communication via Differentiable Discrete Communication Learning
Aditya Kapoor, Yash Bhisikar, Benjamin Freed +2
Effective communication in multi-agent reinforcement learning (MARL) is critical for success but constrained by bandwidth, yet past approaches have been limited to complex gating m…
Bayesian Inverse Physics for Neuro-Symbolic Robot Learning
Octavio Arriaga, Rebecca Adam, Melvin Laux +4
Real-world robotic applications, from autonomous exploration to assistive technologies, require adaptive, interpretable, and data-efficient learning paradigms. While deep learning…
Deep Reinforcement Learning Agents are not even close to Human Intelligence
Quentin Delfosse, Jannis Blüml, Fabian Tatai +6
Deep reinforcement learning (RL) agents achieve impressive results in a wide variety of tasks, but they lack zero-shot adaptation capabilities. While most robustness evaluations fo…
Inverse decision-making using neural amortized Bayesian actors
Dominik Straub, Tobias F. Niehues, Jan Peters +1
Bayesian observer and actor models have provided normative explanations for many behavioral phenomena in perception, sensorimotor control, and other areas of cognitive science and…