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.

• Type: Journal
• Year: 2010
• Where: JMLR
• Authors: Steve Hanneke, Liu Yang
• Keywords: Convex Loss Functions, JMLR 2010, Passive Learning, Proper Active Learning