(Source: wikipedia.com)

Investigators 
Truyen Tran (Australia) 
Hieu-Chi Dam (Japan) 
Alejandro Franco (France) 

Members  
Kien Do 
Phuoc Nguyen 
Dung Nguyen  
Thang Nguyen  
Kha Pham  

 Alumni  
Adam Beykikhoshk 
Shivapratap Gopakumar 
Vuong Le  
Tu Nguyen  
Trang Pham  

Generative AI for Materials Discovery

Targeted exploration of unvisited regions in the materials space.

Overview:

This project will develop a general Generative AI platform to accelerate inverse design of materials. More specifically, we will:

• Establish a multimodal agent flow to orchestrate scientific tools for automated materials discovery.
• Develop scientific LLMs that learn materials compositions, structures, and properties from millions of scientific texts and materials databases.
• Invent physics-guided [reinforcement learning-based] generative models to conditionally sample novel materials with desired functions.
• Invent a closed-loop optimisation framework which integrates generative models with expert judgement and first-principles calculations, where the AI system iteratively proposes candidates and learns from feedback.
• Demonstrate AI-accelerated discovery of functional materials with far fewer experiments than traditional approaches.
  
• Validate the AI-generated materials designs through high-throughput simulation, and experimental synthesis and characterisation.
• Release open-source software tools and DFT-validated materials database.
  

Talks/Tutorials

• Generative AI to accelerate discovery of materials, Keynote @PRICM11, Nov 2023.
• AI for automated materials discovery via learning to represent, predict, generate and explain, @Thuyloi University, May 2023.
  
• Machine learning and reasoning for drug discovery Tutorial @ECML-PKDD, Sept 2021.
• Modern AI for drug discovery, VietAI Summit, Nov 2019.
• AI for matters, Phenikaa University, Hanoi, Vietnam, Jan 2019.
• Deep learning for biomedicine: Genomics and Drug design, Institute of Big Data, Hanoi, Vietnam, Jan 2019.

Publications

• Enabling discovery of materials through enhanced generalisability of deep learning models, Tawfik, Sherif Abdulkader, Tri Minh Nguyen, Salvy P. Russo, Truyen Tran, Sunil Gupta, and Svetha Venkatesh.  arXiv preprint arXiv:2402.10931.
• Towards understanding structure–property relations in materials with interpretable deep learning,  Tien-Sinh Vu, Minh-Quyet Ha, Duong Nguyen Nguyen, Viet-Cuong Nguyen, Yukihiro Abe, Truyen Tran, Huan Tran, Hiori Kino, Takashi Miyake, Koji Tsuda, Hieu-Chi Dam, npj Computational Materials, 9(215), (2023).
• Hierarchical GFlowNet for crystal structure generation, Nguyen, Tri, Sherif Tawfik, Truyen Tran, Sunil Gupta, Santu Rana, and Svetha Venkatesh. In AI for Accelerated Materials Design-NeurIPS 2023 Workshop. 2023.
• Machine learning-aided exploration of ultrahard materials, Tawfik, Sherif Abdulkader, Phuoc Nguyen, Truyen Tran, Tiffany R. Walsh, and Svetha Venkatesh. The Journal of Physical Chemistry C 126, no. 37 (2022): 15952-15961.
• Learning to discover medicines, Nguyen, Minh-Tri, Thin Nguyen, and Truyen Tran. International Journal of Data Science and Analytics (2022): 1-16.
• Mitigating cold-start problems in drug-target affinity prediction with interaction knowledge transferring, Nguyen, Tri Minh, Thin Nguyen, and Truyen Tran. Briefings in Bioinformatics 23, no. 4 (2022): bbac269.
• Explaining black box drug target prediction through model agnostic counterfactual samples, Nguyen, Tri Minh, Thomas P. Quinn, Thin Nguyen, and Truyen Tran. IEEE/ACM Transactions on Computational Biology and Bioinformatics (2022).
• GEFA: Early fusion approach in drug-target affinity prediction, Tri Minh Nguyen, Thin Nguyen, Thao Minh Le, Truyen Tran, IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2021.
• Variational hyper-encoding networks, P Nguyen, T Tran, S Gupta, S Rana, HC Dam, S Venkatesh, ECML-PKDD'21, 2021
• Variational hyper-encoding networks, P Nguyen, T Tran, S Gupta, S Rana, HC Dam, S Venkatesh, ECML-PKDD'21, 2021
• Toward a generalization metric for deep generative models, Thanh-Tung, Hoang, and Truyen Tran. NeurNIPS 2020 1st Workshop on I Can’t Believe It’s Not Better.
• GEFA: Early Fusion Approach in Drug-Target Affinity Prediction, Tri Minh Nguyen, Thin Nguyen, Thao Minh Le, Truyen Tran, Machine Learning for Structural Biology (MLSB) Workshop at NeurIPS 2020.
• HyperVAE: A minimum description length variational hyper-encoding network, Phuoc Nguyen, Truyen Tran, Sunil Gupta, Santu Rana, Hieu-Chi Dam, Svetha Venkatesh, NeurIPS 2020 Workshop on Meta-Learning
• On catastrophic forgetting and mode collapse in Generative Adversarial Networks, Thanh-Tung, Hoang, and Truyen Tran, IJCNN'20
  
• Theory and evaluation metrics for learning disentangled representations, K Do, T Tran, ICLR'20.
• Incomplete conditional density estimation for fast materials discovery, Phuoc Nguyen, Truyen Tran, Sunil Gupta, Svetha Venkatesh. SDM'19.
• Graph transformation policy network for chemical reaction prediction, Kien Do, Truyen Tran, Svetha Venkatesh, KDD'19.
• Attentional multilabel learning over graphs: A message passing approach, K Do, T Tran, T Nguyen, S Venkatesh, Machine Learning, 2019.
• Committee machine that votes for similarity between materials; Duong-Nguyen Nguyen, Tien-Lam Pham, Viet-Cuong Nguyen, Tuan-Dung Ho, Truyen Tran, Keisuke Takahashi and Hieu-Chi Dam. IUCrJ, 2018 Nov 1; 5(Pt 6): 830–840.
• Graph memory networks for molecular activity prediction, Trang Pham, Truyen Tran, Svetha Venkatesh, ICPR'18.
• Neural reasoning for chemical-chemical interaction. Trang Pham, Truyen Tran, Svetha Venkatesh, NIPS 2018 Workshop on Machine Learning for Molecules and Materials.
• Improving generalization and stability of Generative Adversarial Networks, Hoang Thanh-Tung, Truyen Tran, Svetha Venkatesh, ICLR'19.
  
• Neural reasoning for chemical-chemical interaction. Trang Pham, Truyen Tran, Svetha Venkatesh, NIPS 2018 Workshop on Machine Learning for Molecules and Materials.
  
• Variational memory encoder-decoder, Hung Le, Truyen Tran, Thin Nguyen, Svetha Venkatesh, NIPS'18.
  
• On catastrophic forgetting and mode collapse in Generative Adversarial Networks, Hoang Thanh-Tung, Truyen Tran, Svetha Venkatesh; ICML Workshop on Theoretical Foundations and Applications of Deep Generative Models, 2018.
  
• Energy-Based Anomaly Detection for Mixed Data, Kien Do, Truyen Tran, Svetha Venkatesh, Knowledge and Information Systems, 2018.
• Column Networks for Collective Classification, Trang Pham, Truyen Tran, Dinh Phung, Svetha Venkatesh, AAAI'17
• Graph classification via deep learning with virtual nodes Trang Pham, Truyen Tran, Hoa Dam, Svetha Venkatesh, Third Representation Learning for Graphs Workshop (ReLiG 2017).
• Graph-induced restricted Boltzmann machines for document modeling, Tu Dinh Nguyen, Truyen Tran, Dinh Phung, and Svetha Venkatesh, Information Sciences, 2016.
• Outlier Detection on Mixed-Type Data: An Energy-based Approach, Kien Do, Truyen Tran, Dinh Phung, Svetha Venkatesh, International Conference on Advanced Data Mining and Applications (ADMA 2016).
• A deep language model for software code, Hoa Khanh Dam, Truyen Tran and Trang Pham, FSE NL+SE 2016.
• Tensor-variate Restricted Boltzmann Machines, Tu Dinh Nguyen, Truyen Tran, Dinh Phung, and Svetha Venkatesh, AAAI 2015. 
• Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (e-NRBM), Truyen Tran, Tu Dinh Nguyen, Dinh Phung, and Svetha Venkatesh, Journal of Biomedical Informatics, 2015, pii: S1532-0464(15)00014-3. doi: 10.1016/j.jbi.2015.01.012. 
• Thurstonian Boltzmann machines: Learning from multiple inequalities, Truyen Tran, Dinh Phung, and Svetha Venkatesh, In Proc. of 30th International Conference in Machine Learning (ICML’13), Atlanta, USA, June, 2013.
• Learning parts-based representations with Nonnegative Restricted Boltzmann Machine, Tu Dinh Nguyen, Truyen Tran, Dinh Phung, and Svetha Venkatesh, Journal of Machine Learning Research (JMLR) Workshop and Conference Proceedings, Vol. 29, Proc. of 5th Asian Conference on Machine Learning, Nov 2013.
• Latent patient profile modelling and applications with Mixed-Variate Restricted Boltzmann Machine, Tu Dinh Nguyen, Truyen Tran, Dinh Phung, and Svetha Venkatesh,  In Proc. of 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD’13), Gold Coast, Australia, April 2013.
• Learning from Ordered Sets and Applications in Collaborative Ranking, Truyen Tran, Dinh Phung and Svetha Venkatesh, in Proc. of. the 4th Asian Conference on Machine Learning (ACML2012), Singapore, Nov 2012.
• Cumulative Restricted Boltzmann Machines for Ordinal Data Analysis, Truyen Tran, Dinh Phung and Svetha Venkatesh, in Proc. of. the 4th Asian Conference on Machine Learning (ACML2012), Singapore, Nov 2012.
• Mixed-Variate Restricted Boltzmann Machines, Truyen Tran, Dinh Phung and Svetha Venkatesh, in Proc. of. the 3rd Asian Conference on Machine Learning (ACML2011), Taoyuan, Taiwan, Nov 2011.
• Ordinal Boltzmann Machines for Collaborative Filtering. Truyen Tran, Dinh Q. Phung and Svetha Venkatesh. In Proc. of 25th Conference on Uncertainty in Artificial Intelligence, June, 2009, Montreal, Canada. Runner-up for the best paper award.

Programming frameworks

• NVIDIA's Modulus

Datasets

• Materials Project

Key references

• Karniadakis, G. E., Kevrekidis, I. G., Lu, L., Perdikaris, P., Wang, S., & Yang, L. (2021). Physics-informed machine learning. Nature Reviews Physics, 3(6), 422-440.