AI Seminar Series 2021: Yangchen Pan

The AI Seminar is a weekly meeting at the University of Alberta where researchers interested in artificial intelligence (AI) can share their research. Presenters include both local speakers from the University of Alberta and visitors from other institutions. Topics can be related in any way to artificial intelligence, from foundational theoretical work to innovative applications of AI techniques to new fields and problems.

On August 27, Yangchen Pan (Amii researcher and PhD candidate at the University of Alberta) presented “Learning Natural Sparse Representations by Fuzzy Tiling Activation" at the AI Seminar.

Recent work has shown that sparse representations -- where only a small percentage of units are active -- can significantly reduce interference. However, that work relied on relatively complex regularization or meta-learning approaches that have only been used offline in a pre-training phase. Yangchen discusses his work pursuing a direction that achieves sparsity by design rather than learning, designing an activation function that produces sparse representations deterministically by construction, making it more amenable to online training.

Watch the full presentation below:

Keep up-to-date on the AI Seminar Series by signing up for the mailing list.

Learn how Amii advances world-leading artificial intelligence and machine learning research: visit our Research page.