New formulations for active learning

Active learning;Sequential analysis;Stochastic optimization

Ganti Mahapatruni, Ravi Sastry ; Gray, Alexander Computer Science Balcan, Maria-Florina Song, Le Rakhlin, Alexander Zhang, Tong ; Gray, Alexander

University:Georgia Institute of Technology

Department:Computer Science

关键词: Active learning;    Sequential analysis;    Stochastic optimization;   

Others  :  https://smartech.gatech.edu/bitstream/1853/51801/1/GANTIMAHAPATRUNI-DISSERTATION-2014.pdf

美国|英语

来源: SMARTech Repository

【 摘 要 】

In this thesis, we provide computationally efficient algorithms with provable statistical guarantees, for the problem of active learning, by using ideas from sequential analysis. We provide a generic algorithmic framework for active learning in the pool setting, and instantiate this framework by using ideas from learning with experts, stochastic optimization, and multi-armed bandits. For the problem of learning convex combination of a given set of hypothesis, we provide a stochastic mirror descent based active learning algorithm in the stream setting.