Benchmarking Predictive Coding Networks -- Made Simple

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pinchetti, Luca, Qi, Chang, Lokshyn, Oleh, Olivers, Gaspard, Emde, Cornelius, Tang, Mufeng, M'Charrak, Amine, Frieder, Simon, Menzat, Bayar, Bogacz, Rafal, Lukasiewicz, Thomas, Salvatori, Tommaso
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917921814478848
author Pinchetti, Luca
Qi, Chang
Lokshyn, Oleh
Olivers, Gaspard
Emde, Cornelius
Tang, Mufeng
M'Charrak, Amine
Frieder, Simon
Menzat, Bayar
Bogacz, Rafal
Lukasiewicz, Thomas
Salvatori, Tommaso
author_facet Pinchetti, Luca
Qi, Chang
Lokshyn, Oleh
Olivers, Gaspard
Emde, Cornelius
Tang, Mufeng
M'Charrak, Amine
Frieder, Simon
Menzat, Bayar
Bogacz, Rafal
Lukasiewicz, Thomas
Salvatori, Tommaso
contents In this work, we tackle the problems of efficiency and scalability for predictive coding networks (PCNs) in machine learning. To do so, we propose a library, called PCX, that focuses on performance and simplicity, and use it to implement a large set of standard benchmarks for the community to use for their experiments. As most works in the field propose their own tasks and architectures, do not compare one against each other, and focus on small-scale tasks, a simple and fast open-source library and a comprehensive set of benchmarks would address all these concerns. Then, we perform extensive tests on such benchmarks using both existing algorithms for PCNs, as well as adaptations of other methods popular in the bio-plausible deep learning community. All this has allowed us to (i) test architectures much larger than commonly used in the literature, on more complex datasets; (ii)~reach new state-of-the-art results in all of the tasks and datasets provided; (iii)~clearly highlight what the current limitations of PCNs are, allowing us to state important future research directions. With the hope of galvanizing community efforts towards one of the main open problems in the field, scalability, we release code, tests, and benchmarks. Link to the library: https://github.com/liukidar/pcx
format Preprint
id arxiv_https___arxiv_org_abs_2407_01163
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Benchmarking Predictive Coding Networks -- Made Simple
Pinchetti, Luca
Qi, Chang
Lokshyn, Oleh
Olivers, Gaspard
Emde, Cornelius
Tang, Mufeng
M'Charrak, Amine
Frieder, Simon
Menzat, Bayar
Bogacz, Rafal
Lukasiewicz, Thomas
Salvatori, Tommaso
Machine Learning
Computer Vision and Pattern Recognition
I.2.6
In this work, we tackle the problems of efficiency and scalability for predictive coding networks (PCNs) in machine learning. To do so, we propose a library, called PCX, that focuses on performance and simplicity, and use it to implement a large set of standard benchmarks for the community to use for their experiments. As most works in the field propose their own tasks and architectures, do not compare one against each other, and focus on small-scale tasks, a simple and fast open-source library and a comprehensive set of benchmarks would address all these concerns. Then, we perform extensive tests on such benchmarks using both existing algorithms for PCNs, as well as adaptations of other methods popular in the bio-plausible deep learning community. All this has allowed us to (i) test architectures much larger than commonly used in the literature, on more complex datasets; (ii)~reach new state-of-the-art results in all of the tasks and datasets provided; (iii)~clearly highlight what the current limitations of PCNs are, allowing us to state important future research directions. With the hope of galvanizing community efforts towards one of the main open problems in the field, scalability, we release code, tests, and benchmarks. Link to the library: https://github.com/liukidar/pcx
title Benchmarking Predictive Coding Networks -- Made Simple
topic Machine Learning
Computer Vision and Pattern Recognition
I.2.6
url https://arxiv.org/abs/2407.01163