A rate-distortion based switching strategy for adaptive state-action abstractions in RL decomposes value error into Bellman residual and bisimulation metric terms to achieve near-optimal performance under lossy compression in tabular settings.
Scalable Methods for Computing State Simi- larity in Deterministic Markov Decision Processes, in: Proceedings of the AAAI Conference on Artificial Intelligence, pp
3 Pith papers cite this work, alongside 10 external citations. Polarity classification is still indexing.
3
Pith papers citing it
10
external citations · Crossref
fields
cs.LG 3years
2026 3representative citing papers
A VAE-based latent task representation enables automatic curriculum generation in CRL for non-Euclidean navigation tasks, outperforming interpolation and GAN-based methods in experiments.
In a tabular Q-learning agent navigating Voronoi-derived graphs, task similarity alone shows no statistically significant effect on catastrophic forgetting, with high variability and a possible interaction with task complexity.
citing papers explorer
-
Adaptive state-action abstractions via rate-distortion
A rate-distortion based switching strategy for adaptive state-action abstractions in RL decomposes value error into Bellman residual and bisimulation metric terms to achieve near-optimal performance under lossy compression in tabular settings.
-
Curriculum reinforcement learning with measurable task representation learning
A VAE-based latent task representation enables automatic curriculum generation in CRL for non-Euclidean navigation tasks, outperforming interpolation and GAN-based methods in experiments.
-
Catastrophic Forgetting in Continual Reinforcement Learning
In a tabular Q-learning agent navigating Voronoi-derived graphs, task similarity alone shows no statistically significant effect on catastrophic forgetting, with high variability and a possible interaction with task complexity.