Iterative VAE as a predictive brain model for out-of-distribution generalization

12/01/2020
by   Victor Boutin, et al.
0

Our ability to generalize beyond training data to novel, out-of-distribution, image degradations is a hallmark of primate vision. The predictive brain, exemplified by predictive coding networks (PCNs), has become a prominent neuroscience theory of neural computation. Motivated by the recent successes of variational autoencoders (VAEs) in machine learning, we rigorously derive a correspondence between PCNs and VAEs. This motivates us to consider iterative extensions of VAEs (iVAEs) as plausible variational extensions of the PCNs. We further demonstrate that iVAEs generalize to distributional shifts significantly better than both PCNs and VAEs. In addition, we propose a novel measure of recognizability for individual samples which can be tested against human psychophysical data. Overall, we hope this work will spur interest in iVAEs as a promising new direction for modeling in neuroscience.

READ FULL TEXT

page 4

page 11

page 12

research
11/15/2020

Predictive Coding, Variational Autoencoders, and Biological Connections

This paper identifies connections between predictive coding, from theore...
research
05/13/2019

Affine Variational Autoencoders: An Efficient Approach for Improving Generalization and Robustness to Distribution Shift

In this study, we propose the Affine Variational Autoencoder (AVAE), a v...
research
06/30/2021

Applications of the Free Energy Principle to Machine Learning and Neuroscience

In this PhD thesis, we explore and apply methods inspired by the free en...
research
03/01/2023

Implementing engrams from a machine learning perspective: matching for prediction

Despite evidence for the existence of engrams as memory support structur...
research
05/21/2018

The Roles of Supervised Machine Learning in Systems Neuroscience

Over the last several years, the use of machine learning (ML) in neurosc...
research
03/03/2022

Compressed Predictive Information Coding

Unsupervised learning plays an important role in many fields, such as ar...
research
05/28/2018

A Neurobiological Cross-domain Evaluation Metric for Predictive Coding Networks

Achieving a good measure of model generalization remains a challenge wit...

Please sign up or login with your details

Forgot password? Click here to reset