AI Paper Atlas

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

RiskZero: Plan More to Risk Less with a Learned Model

Yousef Yassin, Junfeng Wen

ICML 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Tree search and policy optimization

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
optimize trajectory-level risk with look-ahead search (RiskZero AZ) instead of Sampled TQLSampled TQL · SBM 60-30, ER 10% density0.25-CVaR return: 38.78 -> 56.86Improved
optimize trajectory-level risk with look-ahead search (RiskZero AZ) instead of Sampled TQLSampled TQL · SMIS 60, ER 15% density0.25-CVaR return: 10.58 -> 24.94Improved
use trajectory-level risk instead of naive future-risk optimization (Risk AZ (N) to Risk AZ) (ablation)AlphaZero-style planner · SBM 60-30, ER 15% density0.25-CVaR return: 1.51 -> 57.34Improved
use trajectory-level risk instead of naive future-risk optimization (Risk MZ (N) to Risk MZ) (ablation)MuZero-style planner · SBM 30-10, ER 10% density0.25-CVaR return: 0.0 -> 7.40Improved
add look-ahead search with learned model over Sampled TQLSampled TQL · Mini Grid 0.25-CVaRsamples to optimal policy: 3x fewer samplesImproved
add RiskZero over Sampled TQLSampled TQL · Space Invaders (MinAtar variant) 0.25-CVaRsamples to optimal policy: ~3x more samples for Sampled TQLImproved
use Gumbel planning instead of MuZero heuristic search (ablation)RiskZero · SBM (30,20,15%)episode risk: n=2 simulations exceed all MuZero variants; MuZero needs n>16Improved
increase number of samples D used in search (ablation)RiskZero (Gumbel and MuZero search) · SBM (30,20,15%)episode riskImproved
train with risk-neutral objective, evaluate under 0.25-CVaRRiskZero · Mini Grid0.25-CVaR episode riskHurt

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