AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

Learning to Bet for Horizon-Aware Anytime-Valid Testing

Ege Onur Taga, Samet Oymak, Shubhanshu Shekhar

ICML 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Tree search and policy optimization · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
learn betting policy with universal DQN instead of hedging over ε-greedy schedules or STaR-Betshorizon-aware betting CS · Beta mixture, N=100, mu=0.25CS width at deadline: 0.079 -> 0.066Improved
learn betting policy with DQN instead of STaR-Betshorizon-aware betting CS · Beta mixture, N=100, mu=0.4CS width at deadline: 0.141 -> 0.134Improved
learn betting policy with DQN instead of hedged ε-scheduleshorizon-aware betting CS · Beta mixture, N=100, mu=0.65CS width at deadline: 0.085 -> 0.075Improved
expand DQN action space from 3 to 9 actions (ablation)DQN betting policy · rejection probability by deadlinepowerNo clear effect
use DQN trained on Beta family on real data without fine-tuningDQN betting policy · six real datasets (DNA methylation, humidity), N=100rejection probability by deadline: best in 5/6 casesImproved
replace terminal-power reward with urgency reward (1 - t/N) (ablation)DQN betting policy · Beta mixture, N=100power at t=NHurt
replace terminal-power reward with early-bonus reward 1+(1-t/N) (ablation)DQN betting policy · Beta mixture, N=100power at t=NMixed
apply DQN trained on Beta to logit-normal dataDQN betting policy · logit-normal datapowerImproved
evaluate DQN at longer deadlines N in {250,300,350}DQN betting policy · longer horizonspowerImproved
retrain DQN for alpha=0.01DQN betting policy · alpha=0.01powerImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy