AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

The Utility of Sparse Representations for Control in Reinforcement Learning

Vincent Liu, Raksha Kumaraswamy, Lei Le, Martha White

AAAI 2019 · Reinforcement learning › Visual and representation RL · approach: Learned state and action representations · 55 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 a learned sparse representation rather than a dense neural-network representation for incremental Sarsa controlSarsa(0) · Mountain Car, Puddle World, Acrobot, Catchercontrol performanceImproved
Use tile coding rather than a dense neural-network representationSarsa(0) · Mountain Car, Puddle World, Acrobot, Catchercontrol performanceImproved
Add Set KL instead of exact KL to the ReLU activation regularizer (ablation)ReLU distributionally regularized network · Mountain Car, Puddle World, Acrobot, Catcherinstance sparsityImproved
Add Set KL instead of exact KL to the ReLU activation regularizer (ablation)Sarsa(0) with learned representations · Mountain Car, Puddle World, Acrobot, Catchercontrol performanceImproved
Use ReLU with an exponential target distribution rather than Sigmoid with a Bernoulli target (ablation)Distributionally regularized neural representation · Mountain Car, Puddle World, Acrobot, Catchercontrol performanceImproved
Use ReLU with an exponential target distribution rather than Sigmoid with a Bernoulli target (ablation)Distributionally regularized neural representation · Mountain Car, Puddle Worldactivation overlap: 16.8 vs 256.0; 8.8 vs 256.0Improved
Add an activation ℓ2 regularizer (ablation)Sarsa(0) with learned representations · Mountain Car, Puddle World, Acrobot, Catchercontrol performanceMixed
Add an activation ℓ1 regularizer (ablation)Sarsa(0) with learned representations · Mountain Car, Puddle World, Acrobot, Catchercontrol performanceHurt
Use top-k activation truncation (ablation)Neural representation for Sarsa(0) · Mountain Car, Puddle World, Acrobot, Catchercontrol performanceHurt
Use winner-take-all activation selection across instances (ablation)Neural representation for Sarsa(0) · Mountain Car, Puddle World, Acrobot, Catchercontrol performanceHurt
Add k-sparse truncation to the ReLU+Set-KL network (ablation)ReLU+SKL representation · Mountain Car, Puddle World, Acrobot, Catchercontrol performanceNo clear effect
Use sparse rather than dense representationsSarsa(0) action-value learning · Puddle Worldactivation overlap: 8.8 vs 54.0Improved

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