Making Look-Ahead Active Learning Strategies Feasible with Neural Tangent Kernels

TL;DR

This paper introduces a neural tangent kernel-based method to efficiently approximate look-ahead active learning strategies, enabling scalable sequential data selection and outperforming existing methods on benchmark datasets.

Contribution

It presents a novel approximation technique for look-ahead active learning using neural tangent kernels, reducing computational costs and allowing streaming model updates.

Findings

Outperforms standard look-ahead strategies significantly

Achieves comparable or better results than state-of-the-art methods

Enables sequential active learning without retraining the model after each data point

Abstract

We propose a new method for approximating active learning acquisition strategies that are based on retraining with hypothetically-labeled candidate data points. Although this is usually infeasible with deep networks, we use the neural tangent kernel to approximate the result of retraining, and prove that this approximation works asymptotically even in an active learning setup -- approximating "look-ahead" selection criteria with far less computation required. This also enables us to conduct sequential active learning, i.e. updating the model in a streaming regime, without needing to retrain the model with SGD after adding each new data point. Moreover, our querying strategy, which better understands how the model's predictions will change by adding new data points in comparison to the standard ("myopic") criteria, beats other look-ahead strategies by large margins, and achieves equal or better performance compared to state-of-the-art methods on several benchmark datasets in pool-based active learning.