Disentangling Abstraction from Statistical Pattern Matching in Human and Machine Learning

TL;DR

This paper compares human and neural network performance on abstract tasks, introducing a methodology to distinguish true abstraction learning from statistical pattern matching, revealing humans excel at abstract tasks while neural networks often do not.

Contribution

It presents a novel methodology for creating task metamers to differentiate between abstraction and pattern matching in humans and machines.

Findings

Humans outperform neural networks on abstract tasks.

Neural networks perform better on metamer tasks that match statistical properties.

The study provides a framework for analyzing the nature of learned representations.

Abstract

The ability to acquire abstract knowledge is a hallmark of human intelligence and is believed by many to be one of the core differences between humans and neural network models. Agents can be endowed with an inductive bias towards abstraction through meta-learning, where they are trained on a distribution of tasks that share some abstract structure that can be learned and applied. However, because neural networks are hard to interpret, it can be difficult to tell whether agents have learned the underlying abstraction, or alternatively statistical patterns that are characteristic of that abstraction. In this work, we compare the performance of humans and agents in a meta-reinforcement learning paradigm in which tasks are generated from abstract rules. We define a novel methodology for building "task metamers" that closely match the statistics of the abstract tasks but use a different underlying generative process, and evaluate performance on both abstract and metamer tasks. We find that humans perform better at abstract tasks than metamer tasks whereas common neural network architectures typically perform worse on the abstract tasks than the matched metamers. This work provides a foundation for characterizing differences between humans and machine learning that can be used in future work towards developing machines with more human-like behavior.