Steve Hanneke

8 papers Purdue University

• Agnostic Active Learning Is Always Better Than Passive Learning
• Agnostic Active Learning Is Always Better Than Passive Learning
• Marginal-Nonuniform PAC Learnability
• Non-Uniform Multiclass Learning with Bandit Feedback
• On Union-Closedness of Language Generation
• Optimal Mistake Bounds for Transductive Online Learning
• Optimal Mistake Bounds for Transductive Online Learning
• Tradeoffs between Mistakes and ERM Oracle Calls in Online and Transductive Online Learning