AI & ML
Mixing Regularization for Average-Reward $Q$-Learning and $Q$-Planning
arXiv stat.ML0 views
arXiv:2610.14988v2 Announce Type: replace-cross Abstract: The average-reward setting for reinforcement learning is challenging because, without discounting, the span-contraction of the Bellman operator depends on the mixing properties of the underlying Markov dynamics. Whereas existing work typically assumes contraction, for instance through environment mixing, we study regularization toward a given reward-agnostic, $m$-step mixing policy to control the mixing properties of the learned policy. Our central mixing lemma quantifies how the reference policy's mixing properties transfer to the resulting learned policies. In particular, a single regularizer-dependent quantity simultaneously controls the contraction of the regularized Bellman operator and the exploration of the resulting policies. A second regularizer-dependent quantity controls the bias introduced by regularization, revealing a trade-off between convergence speed and approximation accuracy. Building on this result, we establish finite-time guarantees for regularized $Q$-iteration, synchronous and single-trajectory $Q$-learning, and $Q$-planning. Under one-step mixing, our bounds for approximating the regularized fixed point match the classical model-query bound for relative value iteration and, up to logarithmic factors, an existing sample-complexity bound for synchronous $Q$-learning. We also establish finite-time guarantees under multi-step mixing. For $Q$-planning, we characterize how the planning horizon trades off boundary bias against sample complexity. Finally, we present numerical studies validating the analysis: (i) knowledge of a strong mixing policy improves convergence and exploration, with larger benefits under more severe bottlenecks; and (ii) the observed rankings of $\alpha$-divergence regularizers by bias and convergence speed are consistent with our theoretical analysis.
Source:https://arxiv.org/abs/2610.14988PreviousPragmatic DML with AI-Learned RepresentationsNextInstalace self-hostovaného prohlížeče souborů jako Android APK: jen 64-bit a to je důležité
Related posts
Remove Duplicate Rows in Excel with Java (Compare All or Selected Columns)
0dev.to Java
Your toMap Ran for a Year Until Two Suppliers Used the Same Code
0dev.to Java
Hackers broke into a banking platform without a password and secretly ran programs
0dev.to Java
How to undo a change in Claude Code: /rewind, checkpoints and Git
0dev.to Git
Locking Binary Assets When a Co-Development Team Joins Mid-Project: A Version Control Checklist
0dev.to Git

Real Symlinks in Git Bash on Windows
0dev.to Git