Beyond Disagreement-Based Agnostic Active Learning

Authors: Chicheng Zhang, Kamalika Chaudhuri

NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this paper, we provide such an algorithm. Our solution is based on two key contributions, which may be of independent interest. The first is a general connection between confidence-rated predictors and active learning. ... Our second key contribution is a novel confidence-rated predictor with guaranteed error that applies to any general classification problem. We show that our predictor is optimal in the realizable case, in the sense that it has the lowest abstention rate out of all predictors guaranteeing a certain error. Moreover, we show how to extend our predictor to the agnostic setting. Combining the label query algorithm with our novel confidence-rated predictor, we get a general active learning algorithm consistent in the agnostic setting. We provide a characterization of the label complexity of our algorithm, and show that this is better than the bounds known for disagreementbased active learning in general. Finally, we show that for linear classification with respect to the uniform distribution and log-concave distributions, our bounds reduce to those of [3, 4].
Researcher Affiliation Academia Chicheng Zhang University of California, San Diego 9500 Gilman Drive, La Jolla, CA 92093 chichengzhang@ucsd.edu Kamalika Chaudhuri University of California, San Diego 9500 Gilman Drive, La Jolla, CA 92093 kamalika@cs.ucsd.edu
Pseudocode Yes Algorithm 1 Active Learning Algorithm: Outline; Algorithm 2 An Adaptive Algorithm for Label Query Given Target Excess Error; Algorithm 3 Confidence-rated Predictor
Open Source Code No The paper does not provide any links to open-source code or explicitly state that code for their methodology is available.
Open Datasets No The paper does not refer to specific, publicly available datasets. It discusses theoretical properties related to data distributions but not actual datasets used in experiments.
Dataset Splits No The paper is theoretical and does not describe empirical experiments with dataset splits. Therefore, no validation split information is provided.
Hardware Specification No The paper is theoretical and does not discuss hardware used for experiments.
Software Dependencies No The paper is theoretical and does not specify any software dependencies with version numbers.
Experiment Setup No The paper is theoretical and does not describe an empirical experimental setup with hyperparameters or training configurations.