Predictive Coding Graphs Are a Superset of Feedforward Neural Networks

Abstract

Predictive coding graphs (PCGs) are a recently introduced generalization to predictive coding networks (PCNs), a neuroscience-inspired probabilistic latent variable model. Here, we prove how PCGs define a mathematical superset of feedforward artificial neural networks (multilayer perceptrons). This positions PCNs more strongly within contemporary machine learning, and reinforces earlier proposals to study the use of non-hierarchical neural networks for learning tasks, and more generally the notion of topology in neural networks.

Cite

Text

van Zwol. "Predictive Coding Graphs Are a Superset of Feedforward Neural Networks." NeurIPS 2024 Workshops: NeuroAI, 2024.

Markdown

[van Zwol. "Predictive Coding Graphs Are a Superset of Feedforward Neural Networks." NeurIPS 2024 Workshops: NeuroAI, 2024.](https://mlanthology.org/neuripsw/2024/vanzwol2024neuripsw-predictive/)

BibTeX

@inproceedings{vanzwol2024neuripsw-predictive,
  title     = {{Predictive Coding Graphs Are a Superset of Feedforward Neural Networks}},
  author    = {van Zwol, Björn},
  booktitle = {NeurIPS 2024 Workshops: NeuroAI},
  year      = {2024},
  url       = {https://mlanthology.org/neuripsw/2024/vanzwol2024neuripsw-predictive/}
}