To date, the tightest upper and lower-bounds for the active learning of general concept classes have been in terms of a parameter of the learning problem called the splitting index. We provide, for the first time, an efficient algorithm that is able to realize this upper bound, and we empirically demonstrate its good performance.
Chris Tosh (University of California, San Diego)
Sanjoy Dasgupta (UCSD)
Sanjoy Dasgupta is a Professor in the Department of Computer Science and Engineering at UC San Diego. He works on algorithms for machine learning, with a focus on unsupervised and interactive learning.
Related Events (a corresponding poster, oral, or spotlight)
2017 Talk: Diameter-Based Active Learning »
Mon Aug 7th 08:27 -- 08:45 AM Room C4.8
More from the Same Authors
2018 Tutorial: Understanding your Neighbors: Practical Perspectives From Modern Analysis »
Sanjoy Dasgupta · Samory Kpotufe