Transductive Active Learning: Theory and Applications

Jonas Hübotter, Bhavya Sukhija, Lenart Treven, Yarden As, Andreas Krause

Keywords: active learning, experimental design, bandits, Bayesian optimization, neural networks, deep learning, fine-tuning, transfer learning, transductive learning, generalization, extrapolation

TL;DR: We develop a theory for automatic data selection when you know what you want to learn. We show that knowing what you want a model to learn can be leveraged to learn much more efficiently than just trying to learn "everything".

Abstract:

We study a generalization of classical active learning to real-world settings with concrete prediction targets where sampling is restricted to an accessible region of the domain, while prediction targets may lie outside this region.
We analyze a family of decision rules that sample adaptively to minimize uncertainty about prediction targets.
We are the first to show, under general regularity assumptions, that such decision rules converge uniformly to the smallest possible uncertainty obtainable from the accessible data.
We demonstrate their strong sample efficiency in two key applications: active fine-tuning of large neural networks and safe Bayesian optimization, where they achieve state-of-the-art performance.

Primary Area: Active learning

Submission Number: 2135