AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

GRASP: Navigating Retrosynthetic Planning with Goal-driven Policy

Yemin Yu, Ying Wei, Kun Kuang, Zhengxing Huang, Huaxiu Yao, Fei Wu

NeurIPS 2022 · Reinforcement learning › Visual and representation RL · approach: RL with tree search for discovery · 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
Use GRASP planner instead of Retro*MT single-step planning · Pistachiosuccess rate: 0.92 -> 0.95Improved
Use GRASP planner instead of Retro*MT single-step planning · WuxiTestsuccess rate: 0.80 -> 0.86Improved
Use GRASP planner instead of Retro*MT single-step planning · Pistachioavg. time: 62.0 -> 42.6Improved
Use GRASP planner instead of Retro*MT single-step planning · WuxiTestavg. time: 112.5 -> 79.3Improved
Use GRASP instead of HgSearch on average costMT single-step planning · Pistachioavg. cost: 7.06 -> 7.47Hurt
Use GRASP instead of HgSearch on average costMT single-step planning · WuxiTestavg. cost: 21.41 -> 21.19Improved
Use GRASP instead of Retro* for pathway lengthMT single-step planning · WuxiTestavg. length: 7.50 -> 6.93Improved
Use GRASP instead of Retro* for pathway lengthMT single-step planning · Pistachioavg. length: 4.27 -> 4.12Improved
Specify expert goals for goal-driven planning instead of general planningGRASP · WuxiTest user studyavg. rating (0-10): 8.3 -> 9.2Improved
Use GRASP goal-driven planning with MCTS source building blocksGRASP · WuxiTest user studyavg. rating (0-10): 6.5 -> 7.5Improved
Replace Retro* with GRASP in self-improved retrosynthetic planningSelf-improved planning · Pistachio, WuxiTestsuccess rateImproved
Add self-imitation learning (SIL) (ablation)GRASP training · Pistachiotraining success rateImproved

Later papers that compared with this method

1 later accepted paper reports GRASP as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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