AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Squeezing More from the Stream : Learning Representation Online for Streaming Reinforcement Learning

Nilaksh, Antoine Clavaud, Mathieu Reymond, Francois Rivest, Sarath Chandar

ICML 2026 · Reinforcement learning › Visual and representation RL · approach: Learned state and action representations · 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
add SPR auxiliary lossQRC(lambda) · Atari 26 (40M frames, human normalized)IQM return: 0.03 -> 0.61Improved
add orthogonal SPR gradient projection (orth) (ablation)QRC(lambda)+SPR · Atari 26 (40M frames, human normalized)IQM return: 0.61 -> 0.62Improved
add SPR auxiliary lossstreaming DQN · Atari 26 (40M frames, human normalized)IQM return: 0.12 -> 0.22Improved
add naive SPRStream Q(lambda) · Atari 26 (40M frames, human normalized)IQM return: 0.23 -> 0.13Hurt
add orth2 (orthogonalize SPR against ObGD update) (ablation)Stream Q(lambda)+SPR+orth · Atari 26 (40M frames, human normalized)IQM return: 0.12 -> 0.37Improved
add SPR with orth2Stream Q(lambda) · Atari 26 (100M frames, human normalized)IQM return: 0.44 -> 0.56Improved
add SPRstreaming DQN · MinAtar (5M frames)IQM return: 11.96 -> 32.60Improved
add SPR with orthQRC(lambda) · Octax (20M frames)IQM return: 12.52 -> 23.16Improved
add SPR lossQRC(lambda) encoder · Atari 26 (40M frames) effective rankeffective rank: 42.7 -> 1256.9Improved
add SPR lossstreaming DQN encoder · Atari 26 (40M frames) effective rankeffective rank: 518.7 -> 526.7No clear effect
increase prediction depth K from 1 to 5 (ablation)QRC(lambda)+SPR · Atari 26 (40M frames, human normalized)IQM return: 0.46 -> 0.61Improved
use replay buffer size 5 instead of 1 (ablation)streaming DQN · Atari 26 (40M frames, human normalized)IQM return: 0.12 -> 0.13No 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