Exponential Computational Improvement by Reduction

Channel:
United States Simons Institute for the Theory of Computing
Subscribers:
68,700
Published on ● Video Link: https://www.youtube.com/watch?v=uqdo7d0CQkY


Duration: 29:39
642 views
9

John Langford, Microsoft Research New York
Computational Challenges in Machine Learning
https://simons.berkeley.edu/talks/john-langford-2017-5-3



Other Videos By Simons Institute for the Theory of Computing


2017-05-05The “Tell Me Something New” Model of Computation for Machine Learning
2017-05-05Computationally Tractable and Near Optimal Design of Experiments
2017-05-05Computational Challenges and the Future of ML Panel
2017-05-04The Polytope Learning Problem
2017-05-04System and Algorithm Co-Design, Theory and Practice, for Distributed Machine Learning
2017-05-04Does Computational Complexity Restrict Artificial Intelligence (AI) and Machine Learning?
2017-05-04Computational Efficiency and Robust Statistics
2017-05-04Robust Estimation of Mean and Covariance
2017-05-04Embedding as a Tool for Algorithm Design
2017-05-03The Imitation Learning View of Structured Prediction
2017-05-03Exponential Computational Improvement by Reduction
2017-05-03Machine Learning Combinatorial Optimization Algorithms
2017-05-03A Cost Function for Similarity-Based Hierarchical Clustering
2017-05-03Machine Learning for Healthcare Data
2017-05-02Efficient Distributed Deep Learning Using MXNet
2017-05-02Your Neural Network Can't Learn Mine
2017-05-02Sampling Polytopes: From Euclid to Riemann
2017-05-02Understanding Generalization in Adaptive Data Analysis
2017-05-02On Gradient-Based Optimization: Accelerated, Distributed, Asynchronous and Stochastic
2017-05-01Stochastic Gradient MCMC for Independent and Dependent Data Sources
2017-05-01Scaling Up Bayesian Inference for Big and Complex Data


Tags:
Simons Institute
theoretical computer science
UC Berkeley
Computational Challenges in Machine Learning
theory of computing
computer science
John Langford