AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

Scalable Coordinated Exploration in Concurrent Reinforcement Learning

Maria Dimakopoulou, Ian Osband, Benjamin Van Roy

NeurIPS 2018 · Reinforcement learning › Policy gradient and actor-critic · approach: Regularized scalable deep value learning · 14 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 seed sampling instead of episodic PSRLConcurrent tabular cartpole swing-up · Cartpole swing-upevaluation rewardImproved
Use seed sampling instead of concurrent UCRLConcurrent tabular cartpole swing-up · Cartpole swing-upevaluation rewardImproved
Use seed sampling instead of Thompson resamplingConcurrent tabular cartpole swing-up · Cartpole swing-upevaluation rewardImproved
Use seed LSVI or seed TD instead of Thompson resamplingConcurrent reinforcement learning · Bipolar chainmean reward per agentImproved
Use seed LSVI or seed TD instead of concurrent UCRLConcurrent reinforcement learning · Parallel chainsmean reward per agentImproved
Use seed LSVI or seed TD instead of seed samplingConcurrent reinforcement learning · Bipolar chainmean reward per agentNo clear effect
Use seed LSVI or seed TD instead of seed samplingConcurrent reinforcement learning · Parallel chainsmean reward per agentNo clear effect
Use seed TD-ensemble instead of DQN with epsilon-greedy explorationNeural-network concurrent reinforcement learning · High-dimensional cartpole swing-upaverage reward per agentImproved
Increase the number of parallel agentsSeed TD-ensemble · High-dimensional cartpole swing-upaverage reward per agentImproved
Cap ensemble size at min(K, 30) instead of using a separate model for each agent (ablation)Seed TD-ensemble · High-dimensional cartpole swing-upaverage reward per agentNo 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