Noise-Tolerant Interactive Learning from Pairwise Comparisons

Abstract

We study the problem of interactively learning a binary classifier using noisy labeling and pairwise comparison oracles, where the comparison oracle answers which one in the given two instances is more likely to be positive. Learning from such oracles has multiple applications where obtaining direct labels is harder but pairwise comparisons are easier, and the algorithm can leverage both types of oracles. In this paper, we attempt to characterize how the access to an easier comparison oracle helps in improving the label and total query complexity. We show that the comparison oracle reduces the learning problem to that of learning a threshold function. We then present an algorithm that interactively queries the label and comparison oracles and we characterize its query complexity under Tsybakov and adversarial noise conditions for the comparison and labeling oracles. Our lower bounds show that our label and total query complexity is almost optimal.

28 pages, 1 figure, 3 tables

References in corpus (9)

• Active Ranking using Pairwise Comparisons
• Beyond Disagreement-based Agnostic Active Learning
• When is it Better to Compare than to Score?
• Surrogate Losses in Passive and Active Learning
• Learning Combinatorial Functions from Pairwise Comparisons
• Active Ranking from Pairwise Comparisons and when Parametric Assumptions Don't Help
• Interactive algorithms: from pool to stream
• Search Improves Label for Active Learning
• Active classification with comparison queries