What we believe

"Here's to the crazy ones. The misfits, the rebels, the troublemakers, the round pegs in the square holes, the ones who see things differently. They're not fond of rules, and they have no respect for the status quo. You can quote them, disagree with them, glorify or vilify them. About the only thing you can't do is ignore them. Because they change things - they push the human race forward. And while some may see them as the crazy ones, we see genius. Because the people who are crazy enough to think they can change the world, are the ones who do." --Steve Jobs

"The future depends on some graduate student who is deeply suspicious of everything I have said." -- Geoffrey Hinton

CogAI4Sci team

We are the Cognitive AI for Science (CogAI4Sci) team at the National University of Singapore. We develop creative AI systems that propose novel ideas to drive scientific discovery. Our work focuses on enhancing the creativity of these AI systems, using inspiration from human cognition, and applying them to real-world challenges in biomedical research—supporting human scientists and accelerating breakthroughs.

Before starting CogAI4Sci team, Dianbo Liu was a group leader at the Broad Institute of MIT and Harvard. Prior to the Broad Institute, Dianbo worked as a postdoctoral researcher with Prof. Yoshua Bengio (a Turing Award winner) and led the Humanitarian AI team at the Mila-Quebec AI Institute. This followed his fellowship training and studies in medical informatics at Harvard Medical School. Dianbo earned his PhD from the University of Dundee, Scotland, under the supervision of Prof. Timothea Newman. During his doctoral studies,he received the Vest Scholarship from the Massachusetts Institute of Technology (MIT) and was a special graduate student at the MIT Computer Science and Artificial Intelligence Lab. Dianbo also co-founded two start-ups, "GeneTank" and "SecureAILabs," to advance AI applications in biomedical sciences during his training.

News

• [Oct 2024] Our decoupled straight-through discrete optimization available onine.

• [Oct 2024] Our work on Gaussian mixture continual learning available onine.

• [Oct 2024] Our work on discrete concept unlearning available onine.

• [Oct 2024] Our work on brain inspired continual learning available onine.

• [Sep 2024] Our review on AI safety in medicine available onine.

• [June 2024] Our work on self-supervised learning on medical data BarlowTwins-CXR is published on BMC. Congratulatiosn to Haoyue and all co-authors.

• [Nov 2023] Our exploration of generative models for causal discovery of gene networks Swift-DynGFN is accepted at Neurips Generative model for biology workshop. Congratulatiosn to Trang and all co-authors.

• [Nov 2023] Make our large language model physical reasoning task COAT is available

• [Sep 2023] our work Reusable Slotwise Mechanisms is accepted at Neurips 2023. Congratulatiosn to Trang and all co-authors.

• [July. 2023] Present our GFlowOut work at ICML 2023 in Hawaii

• [June. 2023] Our 2-year effort on attention schema will be presented at Neurips InforCog workshop

• [Apr. 2023] Present our SAF paper at ICLR 2023 at Kigali, Rwanda

Join our team

We have multiple openings for PhD students, postdocs, interns and visiting scholars to join our lab and work with us together in the following directions:

• Creative Automatic Discovery
• Creativity of generative models
• Creative Machine learning for medicine and biology

dianbo at nus dot edu dot sg

Publications

For the most up-to-date list of publications, see my Google Scholar profile.

Selected Machine learning Publications

Selected Machine learning for biomed Publications