AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Reinforcement Learning for Solving the Vehicle Routing Problem

MohammadReza Nazari, Afshin Oroojlooy, Lawrence Snyder, Martin Takac

NeurIPS 2018 · Reinforcement learning › Visual and representation RL · approach: RL for combinatorial routing · 169 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 beam search instead of greedy decoding (ablation)RL · VRP10, Cap20mean tour length: 4.84 -> 4.68Improved
Use beam search instead of greedy decoding (ablation)RL · VRP20, Cap30mean tour length: 6.59 -> 6.40Improved
Use beam search instead of greedy decoding (ablation)RL · VRP50, Cap40mean tour length: 11.39 -> 11.15Improved
Use beam search instead of greedy decoding (ablation)RL · VRP100, Cap50mean tour length: 17.23 -> 16.96Improved
Allow split deliveries by relaxing the demand-versus-load mask (ablation)RL-BS(10) · VRP10, Cap20mean tour length: 4.68 -> 4.65Improved
Allow split deliveries by relaxing the demand-versus-load mask (ablation)RL-BS(10) · VRP20, Cap30mean tour length: 6.40 -> 6.34Improved
Allow split deliveries by relaxing the demand-versus-load mask (ablation)RL-BS(10) · VRP50, Cap40mean tour length: 11.15 -> 11.08Improved
Allow split deliveries by relaxing the demand-versus-load mask (ablation)RL-BS(10) · VRP100, Cap50mean tour length: 16.96 -> 16.86Improved
Remove the embedding layer (ablation)RL · VRPsolution qualityHurt
Add dynamic elements as inputs to the decoder (ablation)RL · VRPsolution qualityNo clear effect
Add an entropy regularizer (ablation)RL training · VRPsolution qualityNo clear effect
Remove the RNN encoder (ablation)Pointer-RL · TSP20average tour length: 3.96 -> 3.97No 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