Algorithms with Machine Learned Advice Two years ago a new research area that uses machine learning to improve classic algorithms and data structures was born. Since then, applications of this approach to indexing, hashing, text indexing, on-line and streaming algorithms […]
Sideways: Depth-Parallel Training of Video Models I will present our recent research on Sideways, an approximate backpropagation scheme for training video models. In standard backpropagation, the gradients and activations at every computation step through the model are temporally synchronized. The […]
Machine learning and theorem proving In my talk I will describe how theorem proving can be phrased as a reinforcement learning problem and provide experimental results with regard to various datasets consisting of mathematical problems ranging from SAT problems to […]
