# Learning Representations by Humans, for Humans

## TL;DR

This paper introduces a novel framework that enhances human decision-making by learning human-facing representations optimized for human performance, integrating human models into the learning process with a unique training method.

## Contribution

It proposes a new framework that reframes machine support as problem reframing, directly optimizing representations for human decision-making rather than just machine accuracy.

## Key findings

- Framework successfully applied to various tasks
- Improves human decision-making performance
- Incorporates human decision models into representation learning

## Abstract

When machine predictors can achieve higher performance than the human decision-makers they support, improving the performance of human decision-makers is often conflated with improving machine accuracy. Here we propose a framework to directly support human decision-making, in which the role of machines is to reframe problems rather than to prescribe actions through prediction. Inspired by the success of representation learning in improving performance of machine predictors, our framework learns human-facing representations optimized for human performance. This "Mind Composed with Machine" framework incorporates a human decision-making model directly into the representation learning paradigm and is trained with a novel human-in-the-loop training procedure. We empirically demonstrate the successful application of the framework to various tasks and representational forms.