AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Fast Approximate Dynamic Programming for Infinite-Horizon Markov Decision Processes

Mohamad Amin Sharifi Kolarijani, Gyula Felix Max, Peyman Mohajerin Esfahani

NeurIPS 2021 · Learning theory and algorithms › Combinatorial and search optimization · approach: Predict-then-optimize learning · 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 ConjVI instead of primal VIvalue iteration · Synthetic stochastic system, N=41iterations to termination: 102 -> 55Improved
use dynamic dual grid (ConjVI-d) instead of static dual grid (ablation)ConjVI · Synthetic stochastic systemiterations to termination: 55 -> 100Hurt
use dynamic dual grid (ConjVI-d) instead of static dual grid (ablation)ConjVI · Synthetic stochastic systemgreedy policy costImproved
use ConjVI instead of primal VIvalue iteration · Synthetic, inverted pendulumgreedy policy average costHurt
use ConjVI instead of primal VIvalue iteration · Batch reactor eta=0.01run-time (sec): 7669 -> 55Improved
use ConjVI instead of primal VIvalue iteration · Batch reactor eta=0.01average cost (100 runs): 33.9 -> 73.5Hurt
use ConjVI-d instead of primal VI when cost has large Lipschitz constantvalue iteration · Batch reactor eta=0.01average cost (100 runs): 33.9 -> 74.0Hurt
use dynamic dual grid instead of static dual grid (ablation)ConjVI · Batch reactor eta=0.01run-time (sec): 55 -> 90Hurt
use dynamic dual grid instead of static dual grid (ablation)ConjVI · Batch reactor (default cost)greedy policy costImproved
deterministic dynamics instead of stochasticConjVI-d · Synthetic systemiterations to termination: 100 -> 10Improved
use analytic conjugate of input cost instead of numerical approximation (ablation)ConjVI · Synthetic systemvalue function and iterationsNo clear effect
scaling factor alpha = 3 for dynamic dual grid (ablation)ConjVI-d · Inverted pendulumconvergenceHurt

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