AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

POMO: Policy Optimization with Multiple Optima for Reinforcement Learning

Yeong-Dae Kwon, Jinho Choo, Byoungjip Kim, Iljoo Yoon, Youngjune Gwon, Seungjai Min

NeurIPS 2020 · Reinforcement learning › Visual and representation RL · approach: Scalable RL systems and environments · 135 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
train with POMO (shared multi-start baseline) instead of REINFORCE greedy rollout baseline (ablation)AM · TSP100optimality gap: 3.51% -> 1.07%Improved
train with POMO instead of REINFORCE (ablation)AM · CVRP100gap: 6.72 -> 3.00Improved
add multiple greedy rollouts from all start nodes at inference (ablation)POMO · TSP100optimality gap: 1.07 -> 0.46Improved
add x8 instance augmentation at inference (ablation)POMO · TSP100optimality gap: 0.46 -> 0.14Improved
add x8 instance augmentation at inference (ablation)POMO · CVRP100gap: 0.98 -> 0.32Improved
add multiple greedy rollouts at inference (ablation)POMO · KP100gap: 0.125 -> 0.006Improved
add x8 augmentation to REINFORCE-trained AM (ablation)AM · TSP100optimality gap: 3.51 -> 1.67Improved

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