Open Problem: Active Representation Learning

TL;DR

This paper introduces Active Representation Learning, a new framework combining exploration and representation learning in partially observable environments, with applications like adaptive microscopy for scientific discovery.

Contribution

It proposes a novel problem class that integrates exploration and representation learning, extending active SLAM ideas to scientific discovery contexts.

Findings

Framework for deriving exploration skills from representations

Application to adaptive microscopy in scientific discovery

Enhanced data collection efficiency in natural sciences

Abstract

In this work, we introduce the concept of Active Representation Learning, a novel class of problems that intertwines exploration and representation learning within partially observable environments. We extend ideas from Active Simultaneous Localization and Mapping (active SLAM), and translate them to scientific discovery problems, exemplified by adaptive microscopy. We explore the need for a framework that derives exploration skills from representations that are in some sense actionable, aiming to enhance the efficiency and effectiveness of data collection and model building in the natural sciences.