AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Learning Finite State Representations of Recurrent Policy Networks

Anurag Koul, Alan Fern, Sam Greydanus

ICLR 2019 · Reinforcement learning › Visual and representation RL · approach: Learned state and action representations · 29 citations (OpenAlex)

Read the paper

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
Insert quantized bottlenecks for recurrent state and observations, then fine-tune when neededAtari RNN policies · Pong, Freeway, Bowling, Boxingscore: Pong 21 -> 21; Freeway 21 -> 21; Bowling 60 -> 60; Boxing 100 -> 100No clear effect
Insert quantized bottlenecks and fine-tuneAtari RNN policies · Breakoutscore: 773 -> 423Hurt
Insert quantized bottlenecks and fine-tuneAtari RNN policies · Space Invadersscore: 1820 -> 1335Hurt
Fine-tune the inserted bottlenecks (ablation)Pong RNN · Pongscore: 20 -> 21Improved
Fine-tune the inserted bottlenecks (ablation)Breakout RNN · Breakoutscore: 32 -> 423Improved
Fine-tune the inserted bottlenecks (ablation)Space Invaders RNN · Space Invadersscore: 520 -> 1335Improved
Fine-tune the inserted bottlenecks (ablation)Boxing RNN · Boxingscore: 94 -> 100Improved
Increase hidden-state bottleneck size from 8 to 16 (ablation)Tomita Grammar 5 MMN · Tomita Grammar 5accuracy: 95 -> 100Improved
Fine-tune the inserted hidden-state bottleneck (ablation)Tomita Grammar 5 MMN with B_h=8 · Tomita Grammar 5accuracy: 95 -> 96Improved
Use B_h=16 instead of B_h=8 (ablation)Blind MCE MMN with B_f=8 · Blindaccuracy: 0.78 -> 1Improved
Use both observation and hidden-state bottlenecks at B_f=4 and B_h=4 (ablation)Tracker MCE MMN · Trackeraccuracy: 1 -> 0.98Hurt
Increase the observation bottleneck size from B_f=100 to B_f=400 (ablation)Breakout MMN with B_h=64 · Breakoutscore: 423 -> 415No 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