Exploring Deep Recurrent Models with Reinforcement Learning for Molecule Design
Problem
Designing small molecules with bespoke properties for drug discovery is hard, and results across computational methods are difficult to compare.
Why it matters
Small-molecule design is central to drug discovery and the development of new medicines.
What the AI did
Applied deep recurrent networks with reinforcement learning (A2C, PPO) to molecular generation and built 19 expert-selected benchmarks as OpenAI Gym environments, expanding training data to ~1.1 million molecules.
Results
not stated in source
Impact
Open-sourced benchmarks aim to enable comparison and innovation in molecular design algorithms.
Record
| Published | 2018 |
|---|---|
| Model or method | Deep recurrent networks with A2C and PPO reinforcement learning |
| Organization | not stated in source |
| Primary source | https://openalex.org/W2786785157 |
| Keywords | reinforcement learning, molecule design, benchmarks, recurrent networks, drug discovery |
Importance rating
significant — thousands to millions affected
Abstract states small-molecule design is of central importance to drug discovery and new medicines; open-sourced benchmarks enable comparison and innovation in molecular design.
More in Science & Research
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Accelerating drug target inhibitor discovery with a deep generative foundation model | 2023 | 51 |
8
|
| 2 | Discovery of senolytics using machine learning | 2023 | 130 |
8
|
| 3 | Deep Learning Pipeline for Accelerating Virtual Screening in Drug Discovery | 2024 | 54 |
8
|
| 4 | Matrix Factorization-based Technique for Drug Repurposing Predictions | 2020 | 56 |
8
|
| 5 | A Systematic Review of Deep Learning Methodologies Used in the Drug Discovery Process with Emphasis on In Vivo Validation | 2023 | 35 |
8
|