AI Paper Atlas

Research map / Learning theory and algorithms / Submodular and game algorithms

The Efficiency of the HyperPlay Technique Over Random Sampling

Michael Schofield, Michael Thielscher

AAAI 2017 · Learning theory and algorithms › Submodular and game algorithms · approach: Zero-sum game learning dynamics · 2 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
apply weight remedy for unbalanced samples (ablation)HyperPlayer · Battleships In Fogwin rate: 82.0 -> 85.3Improved
apply weight remedy for unbalanced samples (ablation)HyperPlayer · Transitwin rate: 72.2 -> 74.6Improved
apply balance remedy for unbalanced samples (ablation)HyperPlayer · Mastermindwin rate: 92.8 -> 93.8No clear effect
apply balance remedy for unbalanced samples (ablation)HyperPlayer · Hidden Connectwin rate: 37.2 -> 36.0No clear effect
apply weight remedy for unbalanced samples (ablation)HyperPlayer · Border Protectionwin rate: 51.0 -> 49.2Hurt
apply either remedy for biased samples (ablation)HyperPlayer · 8 gameswin rateNo clear effect
use HyperPlay backtracking sampling instead of random samplingHyperPlayer · Blind Breakthrough full size, round 32states visited per sampleImproved
use HyperPlay instead of random sampling over a long gameHyperPlayer · cut-down Blind Breakthroughcost of sampling relative to random: 2.2%Improved
move from mid-game to end-game samplingHyperPlay · Transitmedian uniformity probability: 0.001 -> 0.999Improved
move from mid-game to end-game samplingHyperPlay · Blind Breakthroughmedian uniformity probability: 0.156 -> 0.998Improved
move from mid-game to end-game samplingHyperPlay · Hidden Connectmedian uniformity probability: 0.001 -> 0.001No clear effect

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