Research map / Probabilistic and scientific modeling / Molecules and proteins
TorsionNet: A Reinforcement Learning Approach to Sequential Conformer Search
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| use TorsionNet instead of RDKit | conformer generation · 11 torsion alkane | Gibbs Score: 1.05 -> 1.72 | Helped |
| use TorsionNet instead of RDKit on out-of-distribution larger molecule | conformer generation · 22 torsion alkane | Gibbs Score: 1.22 -> 4.75 | Helped |
| use TorsionNet instead of Confab | conformer generation · 11 torsion alkane | Gibbs Score: 0.14 -> 1.72 | Helped |
| use TorsionNet instead of Confab | conformer generation · 22 torsion alkane | Gibbs Score: $\leq10^{-4}$ -> 4.75 | Helped |
| use TorsionNet instead of Confab | conformer generation · 11 torsion alkane | Wall Time (s): 8.87 -> 13.35 | Hurt |
| use TorsionNet instead of RDKit | conformer generation · 22 torsion alkane | Wall Time (s): 68.72 -> 35.23 | Helped |
| use TorsionNet instead of RDKit | conformer generation · 11 torsion alkane | Wall Time (s): 14.52 -> 13.35 | Helped |
| use TorsionNet instead of Confab | conformer generation · 8-lignin | Gibbs Score: $\leq10^{-4}$ -> 0.60 | Helped |
| use TorsionNet instead of enhanced MD (SGMD) | conformer generation · 8-lignin | CPU Time (h): 277.59 -> 0.35 | Helped |
| use TorsionNet instead of enhanced MD (SGMD) | conformer generation · 8-lignin | Gibbs Score: 1.00 -> 0.60 | Hurt |
| use log Gibbs score reward instead of raw Gibbs score | TorsionNet · lignin environment | training stability | Helped |
| train sequentially with curriculum and direct parameter transfer on T-Branched Alkanes | TorsionNet · T-Branched Alkanes | energy distance to best conformer | Helped |
Closest papers by meaning
- Gradual Optimization Learning for Conformational Energy Minimization (ICLR 2024 · similarity 0.89)
- Learning Over Molecular Conformer Ensembles: Datasets and Benchmarks (ICLR 2024 · similarity 0.88)
- Swallowing the Bitter Pill: Simplified Scalable Conformer Generation (ICML 2024 · similarity 0.88)
- DiSCO: Diffusion Schrödinger Bridge for Molecular Conformer Optimization (AAAI 2024 · similarity 0.87)
- Searching for High-Value Molecules Using Reinforcement Learning and Transformers (ICLR 2024 · similarity 0.86)
- Aligning Transformers with Continuous Feedback via Energy Rank Alignment (NeurIPS 2025 · similarity 0.86)
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.