Neural Network Computation
Yann and Keith discuss the challenges of handling variable numbers of layers in neural networks, highlighting the difference between discrete symbolic computation and differentiable optimization. They explore the potential of augmenting neural networks with Turing machines to bridge the capability gap in computation.In this clip
From this podcast
Machine Learning Street Talk (MLST)
061: Interpolation, Extrapolation and Linearisation (Prof. Yann LeCun, Dr. Randall Balestriero)
Related Questions