Low data / Transfer Learning (TL) / few-shot


In chemistry, we often see datasets of <1000 points, and machine learning models built for text/vision applications generally aren't applicable in this low-data regime. Building models that can adapt well to small datasets will be a crucial goal for chemistry machine learning moving forward. In C-CAS, we utilize novel few-shot algorithms and low-data representations to pre-train models on a dataset and then fine-tune on a few available examples of the downstream target task, thus allowing the model to make predictions with minimal training data.


  • Ziyi Kou , Shichao Pei , Yijun Tian and Xiangliang Zhang.  A Controllable Prompt Adversarial Attacking Framework for Black-Box Text Guided Image Generation Models.  2023 Intl. Joint Conf. Art Intel https://www.ijcai.org/proceedings/2023/0109.pdf

  • Guo, Zhichun, Chunhui Zhang, Yujie Fan, Yijun Tian, Chuxu Zhang, and Nitesh V. Chawla. "Boosting graph neural networks via adaptive knowledge distillation." In Proceedings of the AAAI Conference on Artificial Intelligence, 2023 37, 7793-7801. https://doi.org/10.1609/aaai.v37i6.25944

  • Guo, Z., Zhang, C., Yu, W., Herr, J., Wiest, O., and Chawla, N.V. Few-Shot Graph Learning for Molecular Property Prediction. Proc. TheWebConf2021 2021, 2559-2567 https://doi.org/10.1145/3442381.3450112

  • Guo, Z., Yu, W., Zhang, C., Jiang, M. and Chawla, N.V. GraSeq: Graph and Sequence Fusion Learning for Molecular Property Prediction. Proc. 29th ACM Intl. Conf. Inf. Knowl. Manag. 2020, 435-443. https://dl.acm.org/doi/10.1145/3340531.3411981

  • Tang P, Jiang M, Xia BN, Pitera JW, Welser J, Chawla NV. Multi-label patent categorization with non-local attention-based graph convolutional network. Proc. AAAI Conf. Art. Int.  2020 34, 9024-9031.   https://ojs.aaai.org/index.php/AAAI/article/view/6435

  • Saebi, M.;  Nan, B.;  Herr, J.;  Wahlers, J.;  Guo, Z.; Zuranski, A. M.;  Kegej, T.;  Norrby, P.-O.;  Doyle, A. G.;  Wiest, O.; Chawla, N., Wiest, O. On the Use of Real-World Data Sets for Reaction Yield Prediction.  Chem. Sci., 2023, 14, 4997-5005.  https://doi.org/10.1039/D2SC06041H

  • Liu, Zhen, Yurii S. Moroz, and Olexandr Isayev. "The Challenge of Balancing Model Sensitivity and Robustness in Predicting Yields: A Benchmarking Study of Amide Coupling Reactions." Chem.  Sci. 2023, 14, 10835-10846.  doi https://doi.org/10.1039/D3SC03902A