AI Paper Atlas

Research map / Reinforcement learning / MDPs and planning

Reinforcement Learning of Risk-Constrained Policies in Markov Decision Processes

Tomáš Brázdil, Krishnendu Chatterjee, Petr Novotný, Jiří Vahala

AAAI 2020 · Reinforcement learning › MDPs and planning · approach: Robust MDP policy gradient · 4 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
Use RAlph instead of RAMCPHallway 3, Δ=0.1, sim=100 · Hallway 3Avg payoff: 93.18 -> 281.723Improved
Use RAlph instead of RAMCPHallway 3, Δ=0.1, sim=100 · Hallway 3Risk: 0.649 -> 0.001Improved
Use RAlph instead of RAMCPRW 1, Δ=0.05 · RW 1Avg payoff: 17.78 -> 23.32Improved
Use RAlph instead of RAMCPRW 1, Δ=0.05 · RW 1Risk: 0.234 -> 0.035Improved
Increase RAMCP simulations per step from 100 to 500 (ablation)Hallway 3, Δ=0.1 · Hallway 3Avg payoff: 93.18 -> 161.84Improved
Reduce RAlph simulations per step from 100 to 50 (ablation)RAlph · Hallway 3, Δ=0.1Avg payoff: 281.723 -> 279.32No clear effect
Increase the risk threshold from Δ=0 to Δ=0.1 (ablation)RAlph · Hallway 3Avg payoff: 281.169 -> 281.723No clear effect
Use risk-constrained exploration rather than unconstrained explorationRAlph · MDP benchmarkschance of finding promising pathsImproved

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