Duy Nguyen from the Interactive Machine Learning department, along with collaborators from the University of Oldenburg, the University of Texas at Austin, the University of California San Diego, and others, presented a full paper at AAAI 2023. This year’s conference was held in Washington, D.C., USA from February 7th-14th 2023 (overall acceptance rate 19%).
The accepted paper “Joint Self-Supervised Image-Volume Representation Learning with Intra-Inter Contrastive Clustering” is about self-supervised learning (SSL) problems, which offer the ability to overcome the lack of labelled training samples by learning feature representations from unlabeled data. For this, the authors contributed to the SSL medical imaging literature a new framework that efficiently uses numerous unlabeled data types and is flexible with data dimension barriers in medical downstream tasks. For future work, we plan to extend this approach to various self-supervised learning approaches to test its effectiveness and reliability in actual medical scenarios.
The theme of this conference was to create collaborative bridges within and beyond AI. Like previous AAAI conferences, AAAI-23 featured technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and two new activities: a Bridge Program and a Lab Program. Duy reports 4 interesting workshops for IML, Graphs and More Complex Structures for Learning and Reasoning (GCLR); Health Intelligence: Recent Trends in Human-Centric AI: and When Machine Learning Meets Dynamical Systems: Theory and Applications. Our department connected with researchers at IBM Research USA at the last workshop.
