no code implementations • 2 May 2024 • Finn Rietz, Erik Schaffernicht, Stefan Heinrich, Johannes A. Stork
Reinforcement learning policies are typically represented by black-box neural networks, which are non-interpretable and not well-suited for safety-critical domains.
no code implementations • 11 Oct 2023 • Finn Rietz, Johannes Andreas Stork
Discovering all useful solutions for a given task is crucial for transferable RL agents, to account for changes in the task or transition dynamics.
1 code implementation • 3 Oct 2023 • Finn Rietz, Erik Schaffernicht, Stefan Heinrich, Johannes Andreas Stork
PSQD offers the ability to reuse previously learned subtask solutions in a zero-shot composition, followed by an adaptation step.
no code implementations • 20 Sep 2022 • Finn Rietz, Erik Schaffernicht, Todor Stoyanov, Johannes A. Stork
Combining learned policies in a prioritized, ordered manner is desirable because it allows for modular design and facilitates data reuse through knowledge transfer.