Divide-and-Conquer Predictive Coding: a structured Bayesian inference algorithm

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Sennesh, Eli, Wu, Hao, Salvatori, Tommaso
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910653754638336
author Sennesh, Eli
Wu, Hao
Salvatori, Tommaso
author_facet Sennesh, Eli
Wu, Hao
Salvatori, Tommaso
contents Unexpected stimuli induce "error" or "surprise" signals in the brain. The theory of predictive coding promises to explain these observations in terms of Bayesian inference by suggesting that the cortex implements variational inference in a probabilistic graphical model. However, when applied to machine learning tasks, this family of algorithms has yet to perform on par with other variational approaches in high-dimensional, structured inference problems. To address this, we introduce a novel predictive coding algorithm for structured generative models, that we call divide-and-conquer predictive coding (DCPC). DCPC differs from other formulations of predictive coding, as it respects the correlation structure of the generative model and provably performs maximum-likelihood updates of model parameters, all without sacrificing biological plausibility. Empirically, DCPC achieves better numerical performance than competing algorithms and provides accurate inference in a number of problems not previously addressed with predictive coding. We provide an open implementation of DCPC in Pyro on Github.
format Preprint
id arxiv_https___arxiv_org_abs_2408_05834
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Divide-and-Conquer Predictive Coding: a structured Bayesian inference algorithm
Sennesh, Eli
Wu, Hao
Salvatori, Tommaso
Machine Learning
Artificial Intelligence
Neurons and Cognition
Unexpected stimuli induce "error" or "surprise" signals in the brain. The theory of predictive coding promises to explain these observations in terms of Bayesian inference by suggesting that the cortex implements variational inference in a probabilistic graphical model. However, when applied to machine learning tasks, this family of algorithms has yet to perform on par with other variational approaches in high-dimensional, structured inference problems. To address this, we introduce a novel predictive coding algorithm for structured generative models, that we call divide-and-conquer predictive coding (DCPC). DCPC differs from other formulations of predictive coding, as it respects the correlation structure of the generative model and provably performs maximum-likelihood updates of model parameters, all without sacrificing biological plausibility. Empirically, DCPC achieves better numerical performance than competing algorithms and provides accurate inference in a number of problems not previously addressed with predictive coding. We provide an open implementation of DCPC in Pyro on Github.
title Divide-and-Conquer Predictive Coding: a structured Bayesian inference algorithm
topic Machine Learning
Artificial Intelligence
Neurons and Cognition
url https://arxiv.org/abs/2408.05834