Bayesian Active Learning for Classification and Preference Learning

by   Neil Houlsby, et al.

Information theoretic active learning has been widely studied for probabilistic models. For simple regression an optimal myopic policy is easily tractable. However, for other tasks and with more complex models, such as classification with nonparametric models, the optimal solution is harder to compute. Current approaches make approximations to achieve tractability. We propose an approach that expresses information gain in terms of predictive entropies, and apply this method to the Gaussian Process Classifier (GPC). Our approach makes minimal approximations to the full information theoretic objective. Our experimental performance compares favourably to many popular active learning algorithms, and has equal or lower computational complexity. We compare well to decision theoretic approaches also, which are privy to more information and require much more computational time. Secondly, by developing further a reformulation of binary preference learning to a classification problem, we extend our algorithm to Gaussian Process preference learning.


page 1

page 2

page 3

page 4


Characterizing the robustness of Bayesian adaptive experimental designs to active learning bias

Bayesian adaptive experimental design is a form of active learning, whic...

Active Classification: Theory and Application to Underwater Inspection

We discuss the problem in which an autonomous vehicle must classify an o...

OFAI-UKP at HAHA@IberLEF2019: Predicting the Humorousness of Tweets Using Gaussian Process Preference Learning

Most humour processing systems to date make at best discrete, coarse-gra...

Unifying Approaches in Data Subset Selection via Fisher Information and Information-Theoretic Quantities

The mutual information between predictions and model parameters – also r...

Active learning for efficiently training emulators of computationally expensive mathematical models

An emulator is a fast-to-evaluate statistical approximation of a detaile...

GPIRT: A Gaussian Process Model for Item Response Theory

The goal of item response theoretic (IRT) models is to provide estimates...

Predictive Active Set Selection Methods for Gaussian Processes

We propose an active set selection framework for Gaussian process classi...