Valiant's learnability model revisited with new query-based algorithm

Researchers have revisited Valiant's original 1984 learnability model, which differs from the more common PAC learning model by providing only positive examples and allowing membership queries. They established a new characterization for learnability in this model, showing it is strictly between PAC learning and a variant without queries. The study also presents the first algorithm for learning halfspaces within Valiant's model, demonstrating their learnability with queries. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Uncovers theoretical nuances in learnability, potentially influencing future algorithm design for specific problem classes.

RANK_REASON Academic paper analyzing a theoretical model of machine learning. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Valiant
• PAC learning model

• paper
• other