AI Paper Atlas

Research map / Optimization / Nonconvex and smooth optimization

Convergence of Fast Policy Iteration in Markov Games and Robust MDPs

Keith Badger, Jefferson Huang, Marek Petrik

AAAI 2026 · Optimization › Nonconvex and smooth optimization · approach: Equilibrium learning in zero-sum games · 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
use RCPI instead of value iterationpolicy iteration for MGs and RMDPs · Markov games (large)median runtime (s): 253.0 -> 2.3Improved
use RCPI-infinity instead of value iterationpolicy iteration for MGs and RMDPs · Inventory RMDP (large)median runtime (s): 23629.6 -> 84.8Improved
use RCPI-infinity instead of value iterationpolicy iteration for MGs and RMDPs · Gambler's Ruin RMDP (large)median runtime (s): 106.6 -> 54.3Improved
add recovery steps to RCPI (m unbounded vs m=0) (ablation)policy iteration for MGs and RMDPs · Inventory RMDP (large)median runtime (s): 87.8 -> 84.8No clear effect
use RCPI instead of Filar-Tolwinskipolicy iteration for MGs and RMDPs · Gambler's Ruin RMDP (large)median runtime (s): 77.1 -> 54.3Improved
use RCPI instead of PAIpolicy iteration for MGs and RMDPs · Inventory problemsBellman residualImproved
use RCPI instead of Hoffman-Karppolicy iteration for MGs and RMDPs · Gambler's Ruin RMDP (small)median runtime (s): 14.4 -> 4.7Improved
use RCPI instead of Winnicki-Srikantpolicy iteration for MGs and RMDPs · Markov Games (small)median runtime (s): 1.0 -> 0.3Improved
use WS or VI instead of closed-form evaluationpolicy iteration for MGs and RMDPs · Gambler's Ruin RMDP (small)median runtime (s): RCPI 4.7 vs WS 5.2 vs VI 8.7Mixed
use Armijo line search (FT) instead of pure Newton (PAI)policy iteration for MGs and RMDPs · Markov game of Example 4.1convergenceHurt
use PAI instead of FTpolicy iteration for MGs and RMDPs · Inventory problemsBellman residualMixed

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