SOURCE-LINKED INTELLIGENCE

Regularized Emphatic Temporal-Difference Learning: Stability under Constant Stepsizes

arXiv · AI, language, vision and robotics · article · Sep 14, 2026 · UTC

Emphatic temporal-difference learning (ETD) stabilizes the expected off-policy TD update and changes its projection geometry, but neither property determines constant-stepsize sampled dynamics. We construct an ergodic two-state counterexample in which the ETD mean map contracts while the sampled product has a positive top Lyapunov exponent. Regenerative-cycle analysis separates this sign from the infinite variance of the follow-on trace. We introduce regularized emphatic TD (RETD), a normalized first-order post-shock repair that leaves the trace and importance ratios unchanged, stores the emph

Read original source ↗ Open in workspace

recordType

paper

region

Global