{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/using-rule-based-labels-for-weak-supervised","title":"Using Rule-Based Labels for Weak Supervised Learning: A ChemNet for Transferable Chemical Property Prediction","arxiv_id":"1712.02734","date":"2017-12-07","proceeding":null,"authors":["Garrett B. Goh","Charles Siegel","Abhinav Vishnu","Nathan O. Hodas"],"abstract":"With access to large datasets, deep neural networks (DNN) have achieved\nhuman-level accuracy in image and speech recognition tasks. However, in\nchemistry, data is inherently small and fragmented. In this work, we develop an\napproach of using rule-based knowledge for training ChemNet, a transferable and\ngeneralizable deep neural network for chemical property prediction that learns\nin a weak-supervised manner from large unlabeled chemical databases. When\ncoupled with transfer learning approaches to predict other smaller datasets for\nchemical properties that it was not originally trained on, we show that\nChemNet's accuracy outperforms contemporary DNN models that were trained using\nconventional supervised learning. Furthermore, we demonstrate that the ChemNet\npre-training approach is equally effective on both CNN (Chemception) and RNN\n(SMILES2vec) models, indicating that this approach is network architecture\nagnostic and is effective across multiple data modalities. Our results indicate\na pre-trained ChemNet that incorporates chemistry domain knowledge, enables the\ndevelopment of generalizable neural networks for more accurate prediction of\nnovel chemical properties.","url_abs":"http://arxiv.org/abs/1712.02734v2","url_pdf":"http://arxiv.org/pdf/1712.02734v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"using-rule-based-labels-for-weak-supervised","repo_url":"https://github.com/Yindong-Zhang/GraphConvolutionDrugTargetInteration","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"property-prediction","task_name":"Property Prediction"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1712.02734","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}