Deep Learning and the Brain

Sunday, January 20th: 9:00-9:30 – OPENING REMARKS Session 1: Perceptual Representations 9:30-10:15 – Daniel Yamins, Stanford University – Broadening and deepening the role of Artificial Intelligence in Computational Neuroscience 10:15-11:00 – Daphna Weinshall, The Hebrew University – Old new frontiers in visual object recognition using deep learning: curriculum learning 11:00-11:30 – COFFEE BREAK 11:30-12:15 – Kalanit Grill-Spector, Stanford University – The functional neuroanatomy of face perception: from brain measurements to deep neural networks 12:15-13:00 – Adi Mizrahi, The Hebrew University – Perceptual learning in a mouse model: a progress report 13:00-14:00 – LUNCH BREAK Session 2: Theory 14:00-14:45 – Shai Shalev-Shwartz, The Hebrew University – Decoupling gating from linearity 14:45-15:30 – Andrew Saxe, University of Oxford – High-dimensional dynamics of generalization error in neural networks: implications for experience replay 15:30-16:00 – COFFEE BREAK 16:00-16:45 – Naftali Tishby, The Hebrew University – The computational benefit of the hidden layers in Deep Neural Networks 16:45-17:30 – Tomaso Poggio, MIT – Three puzzles in the theory of deep learning 17:30-18:00 – Panel discussion