论文信息 - Negative Results for Active Learning with Convex Losses

Negative Results for Active Learning with Convex Losses

We study the problem of active learning with convex loss functions. We prove that even under bounded noise constraints, the minimax rates for proper active learning are often no better than passive learning.

Liu Yang | Steve Hanneke

[1] Sanjoy Dasgupta,et al. Coarse sample complexity bounds for active learning , 2005, NIPS.

[2] R. Nowak,et al. Upper and Lower Error Bounds for Active Learning , 2006 .

[3] Steve Hanneke,et al. A bound on the label complexity of agnostic active learning , 2007, ICML '07.

[4] Daphne Koller,et al. Support Vector Machine Active Learning with Applications to Text Classification , 2000, J. Mach. Learn. Res..

[5] Steve Hanneke,et al. Teaching Dimension and the Complexity of Active Learning , 2007, COLT.

[6] Steve Hanneke,et al. Adaptive Rates of Convergence in Active Learning , 2009, COLT.

[7] Michael I. Jordan,et al. Convexity, Classification, and Risk Bounds , 2006 .

[8] Sanjoy Dasgupta,et al. A General Agnostic Active Learning Algorithm , 2007, ISAIM.

[9] John Langford,et al. Importance weighted active learning , 2008, ICML '09.

[10] John Langford,et al. Agnostic active learning , 2006, J. Comput. Syst. Sci..

[11] Maria-Florina Balcan,et al. The true sample complexity of active learning , 2010, Machine Learning.

[12] Steve Hanneke,et al. Theoretical foundations of active learning , 2009 .