计算机科学
水准点(测量)
一套
人工智能
超参数
任务(项目管理)
突出
机器学习
公共化学
集合(抽象数据类型)
自然语言处理
过程(计算)
基础(证据)
多样性(控制论)
程序设计语言
生物化学
化学
管理
考古
大地测量学
经济
历史
地理
作者
Walid Ahmad,Elana P. Simon,Seyone Chithrananda,Gabriel Grand,Bharath Ramsundar
出处
期刊:Cornell University - arXiv
日期:2022-01-01
被引量:56
标识
DOI:10.48550/arxiv.2209.01712
摘要
Large pretrained models such as GPT-3 have had tremendous impact on modern natural language processing by leveraging self-supervised learning to learn salient representations that can be used to readily finetune on a wide variety of downstream tasks. We investigate the possibility of transferring such advances to molecular machine learning by building a chemical foundation model, ChemBERTa-2, using the language of SMILES. While labeled data for molecular prediction tasks is typically scarce, libraries of SMILES strings are readily available. In this work, we build upon ChemBERTa by optimizing the pretraining process. We compare multi-task and self-supervised pretraining by varying hyperparameters and pretraining dataset size, up to 77M compounds from PubChem. To our knowledge, the 77M set constitutes one of the largest datasets used for molecular pretraining to date. We find that with these pretraining improvements, we are competitive with existing state-of-the-art architectures on the MoleculeNet benchmark suite. We analyze the degree to which improvements in pretraining translate to improvement on downstream tasks.
科研通智能强力驱动
Strongly Powered by AbleSci AI