Formal Models of Active Learning from Contrastive Examples
Not on the map. Why?
Abstract (arXiv)
Machine learning can greatly benefit from providing learning algorithms with pairs of contrastive training examples -- typically pairs of instances that differ only slightly, yet have different class labels. Intuitively, the difference in the instances helps explain the difference in the class labels. This paper proposes a theoretical framework in which the effect of various types of contrastive examples on active learners is studied formally. The focus is on the sample complexity of learning concept classes and how it is influenced by the choice of contrastive examples. We illustrate our results with geometric concept classes and classes of Boolean functions. Interestingly, we reveal a connection between learning from contrastive examples and the classical model of self-directed learning.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 12 of its 30 references that are in the map. First posted on arXiv 18 Jun 2025.
Builds on
12 of 30 references.
References · 12
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
- 2021Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Losscited by 313
- 2021Investigating the Role of Negatives in Contrastive Representation Learningcited by 54
- 2021Counterfactual State Explanations for Reinforcement Learning Agents via Generative Deep Learningcited by 39
- 2020ALICE: Active Learning with Contrastive Natural Language Explanationscited by 36
- 2018Teaching Categories to Human Learners with Visual Explanationscited by 30
- 2023GANterfactual-RL: Understanding Reinforcement Learning Agents' Strategies through Visual Counterfactual Explanationscited by 17
- 2023Optimal Sample Complexity of Contrastive Learningcited by 15
- 2021Teaching an Active Learner with Contrastive Examplescited by 12
- 2020Robust Learning from Discriminative Feature Feedbackcited by 7
- 2014On Exact Learning Monotone DNF from Membership Queriescited by 6
- 2024The Dimension of Self-Directed Learningcited by 4
Built on by
Citing papers per year: 2026: 2.
2 papers.