Neural representational geometry underlies few-shot concept learning

Ben Sorscher, Surya Ganguli, Haim Sompolinsky*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

31 Scopus citations

Abstract

Understanding the neural basis of the remarkable human cognitive capacity to learn novel concepts from just one or a few sensory experiences constitutes a fundamental problem. We propose a simple, biologically plausible, mathematically tractable, and computationally powerful neural mechanism for few-shot learning of naturalistic concepts. We posit that the concepts that can be learned from few examples are defined by tightly circumscribed manifolds in the neural firing-rate space of higher-order sensory areas. We further posit that a single plastic downstream readout neuron learns to discriminate new concepts based on few examples using a simple plasticity rule. We demonstrate the computational power of our proposal by showing that it can achieve high few-shot learning accuracy on natural visual concepts using both macaque inferotemporal cortex representations and deep neural network (DNN) models of these representations and can even learn novel visual concepts specified only through linguistic descriptors. Moreover, we develop a mathematical theory of few-shot learning that links neurophysiology to predictions about behavioral outcomes by delineating several fundamental and measurable geometric properties of neural representations that can accurately predict the few-shot learning performance of naturalistic concepts across all our numerical simulations. This theory reveals, for instance, that high-dimensional manifolds enhance the ability to learn new concepts from few examples. Intriguingly, we observe striking mismatches between the geometry of manifolds in the primate visual pathway and in trained DNNs. We discuss testable predictions of our theory for psychophysics and neurophysiological experiments.

Original languageEnglish
Article numbere2200800119
JournalProceedings of the National Academy of Sciences of the United States of America
Volume119
Issue number43
DOIs
StatePublished - 25 Oct 2022

Bibliographical note

Publisher Copyright:
Copyright © 2022 the Author(s). Published by PNAS.

Keywords

  • few-shot learning
  • neural networks
  • population coding
  • ventral visual stream

Fingerprint

Dive into the research topics of 'Neural representational geometry underlies few-shot concept learning'. Together they form a unique fingerprint.

Cite this