arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,341 papers · 148 categories

Trend · papers per month

count % of corpus

1 result for “Lazy-PSRL”

Posterior Sampling for Reinforcement Learning Without Episodesstat.ML

Clarifies PSRL for infinite horizon problems without fixed episodes.

problem Applying PSRL in environments without fixed episodes.

method Review and critique of existing results and algorithms.

result Suggests PSRL can achieve

O(\sqrt{T})

bounds under certain assumptions.

PSRL Reinforcement Learning Infinite Horizon Lazy-PSRL Counterexample

2016-08-09abs ↗pdf ↗