Predictive Coding beyond Correlations

Tommaso Salvatori, Luca Pinchetti, Amine M'Charrak, Beren Millidge, Thomas Lukasiewicz

Primary Area: applications to neuroscience & cognitive science

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: Cognitive Science, Bayesian Networks, Predictive Coding

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

Abstract:

Bayesian and causal inference are fundamental processes for intelligence. Bayesian inference models observations: what can be inferred about $y$ if we observe a related variable $x$? Causal inference models interventions: if we directly change $x$, how will $y$ change? Predictive coding is a neuroscience-inspired method for performing Bayesian inference on continuous state variables using local information only.
In this work, we show how a simple change in the inference process of predictive coding enables interventional and counterfactual inference in scenarios where the causal graph is known. We then extend our results, and show how predictive coding can be used in cases where the graph is unknown, and has to be inferred from observational data. This allows us to perform structure learning and causal query answering on predictive coding-based structural causal models. Empirically, we test our method on a large number of benchmarks, as well as presenting experiments that show potential applications in machine learning.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 7655