Timezone: »
Massive scale, both in terms of data availability and computation, enables significant breakthroughs in key application areas of deep learning such as natural language processing (NLP) and computer vision. There is emerging evidence that scale may be a key ingredient in scientific deep learning, but the importance of physical priors in scientific domains makes the strategies and benefits of scaling uncertain. Here, we investigate neural scaling behavior in large chemical models by varying model and dataset sizes over many orders of magnitude, studying models with over one billion parameters, pre-trained on datasets of up to ten million datapoints. We consider large language models for generative chemistry and graph neural networks for machine-learned interatomic potentials. To enable large-scale scientific deep learning studies under resource constraints, we develop the Training Performance Estimation (TPE) framework to reduce the costs of scalable hyperparameter optimization by up to 90%. Using this framework, we discover empirical neural scaling relations for deep chemical models and investigate the interplay between physical priors and scale. Potential applications of large, pre-trained models for "prompt engineering" and unsupervised representation learning of molecules are shown.
Author Information
Connor Coley (MIT)
Nathan C. Frey (Massachusetts Institute of Technology)
More from the Same Authors
-
2022 : Reinforced Genetic Algorithm for Structure-based Drug Design »
Tianfan Fu · Wenhao Gao · Connor Coley · Jimeng Sun -
2022 : Sample Efficiency Matters: Benchmarking Molecular Optimization »
Wenhao Gao · Tianfan Fu · Jimeng Sun · Connor Coley -
2022 Workshop: AI for Science »
Yuanqi Du · Tianfan Fu · Wenhao Gao · Kexin Huang · Shengchao Liu · Ziming Liu · Hanchen Wang · Connor Coley · Le Song · Linfeng Zhang · Marinka Zitnik -
2021 Poster: Non-Autoregressive Electron Redistribution Modeling for Reaction Prediction »
Hangrui Bi · Hengyi Wang · Chence Shi · Connor Coley · Jian Tang · Hongyu Guo -
2021 Spotlight: Non-Autoregressive Electron Redistribution Modeling for Reaction Prediction »
Hangrui Bi · Hengyi Wang · Chence Shi · Connor Coley · Jian Tang · Hongyu Guo -
2020 Poster: Learning to Navigate The Synthetically Accessible Chemical Space Using Reinforcement Learning »
Sai Krishna Gottipati · Boris Sattarov · Sufeng Niu · Yashaswi Pathak · Haoran Wei · Shengchao Liu · Shengchao Liu · Simon Blackburn · Karam Thomas · Connor Coley · Jian Tang · Sarath Chandar · Yoshua Bengio