A Historical View of Large Margin Optimization Methods

Channel:
Microsoft Research
Subscribers:
344,000
Published on ● Video Link: https://www.youtube.com/watch?v=HD_uQeO2jqQ


Duration: 1:02:18
46 views
0

Starting from the design of robust Hopfield works in the early 1990s to SVMs to structured output problems, a variety of methods have been proposed for the solution of optimization problems arising in large margin training. Some of these methods are for the dual while others are for the primal. This talk will give a historical view of these methods, while describing the motivations that led to their development.



Other Videos By Microsoft Research


2016-08-11How To Write a Paper to Optimize its Chances for Publication Success, Part 1
2016-08-11Linear-Algebraic List Decoding And Subspace-Evasive Sets
2016-08-11High Dimensional Data
2016-08-11The Wonders of the Probabilistic Method
2016-08-11Concurrent Data Representation Synthesis
2016-08-11A Domain Specific Language for Testing Concurrent Programs
2016-08-11Planning Under Uncertainty: Challenges and Recent Progress
2016-08-11The Changing Landscape of Parallel Computing - Applications
2016-08-11Summer Number Theory Day; Session 3
2016-08-11Annotating Images with Words, Phrases and Sentences
2016-08-11A Historical View of Large Margin Optimization Methods
2016-08-11Component Based Models: Graphical Models, Sparsity, Low-rank, and all of that Sort of Thing
2016-08-11Adaptive Sampling for Ranking and Clustering
2016-08-11Program Verification via SVMs
2016-08-11Monte Carlo Methods for Bayesian Reinforcement Learning and POMDP
2016-08-11Faculty Summit 2012 - Design Expo
2016-08-11Statistical Consistency and Regret Bounds for Ranking
2016-08-11Achieving High Data Rates in a Distributed MIMO System
2016-08-11Building Knowledge Bases from the Web
2016-08-11Random Graph Models of Kidney Exchange
2016-08-11Useful Spatio-Temporal Abstractions in Reinforcement Learning?


Tags:
microsoft research