AI ALIGNMENT FORUM
AF

1043
Wikitags

Predictive Processing

Edited by Yoav Ravid, Ariel Cheng, Roman Leventov last updated 29th Oct 2025

(From Wikipedia) predictive processing (a.k.a. predictive coding, the Bayesian Brain hypothesis) is a theory of brain function in which the brain is constantly generating and updating a mental (generative) model of the environment. The model is used to generate predictions of sensory input that are compared to actual sensory input. This comparison results in prediction errors that are then used to update and revise the mental model.

Active Inference (AIF) can be seen as a generalization of predictive processing. While predictive processing only explains the agent's perception in terms of inference, AIF models both perception and action as inference under closely related objectives: whereas perception minimizes variational free energy (which sometimes reduces to precision-weighted prediction error), action minimizes expected free energy.

The Free Energy Principle is a generalization of AIF that not only attempts to describe biological organisms, but rather all systems that maintain a separation from their environments (via Markov blankets).

 

External Links:
Book Review: Surfing Uncertainty - Introduction to predictive processing by Scott Alexander
Predictive Processing And Perceptual Control by Scott Alexander
Predictive coding under the free-energy principle by Karl Friston and Stefan Kiebel

 

Related Pages: Perceptual Control Theory, Neuroscience, Free Energy Principle

Subscribe
Discussion
2
Subscribe
Discussion
2
Posts tagged Predictive Processing
17Multi-agent predictive minds and AI alignment
Jan_Kulveit
7y
0
25How uniform is the neocortex?
zhukeepa
5y
1
31Why Simulator AIs want to be Active Inference AIs
Jan_Kulveit, rosehadshar
3y
4
20What Decision Theory is Implied By Predictive Processing?
Q
johnswentworth, abramdemski
5y
Q
12
17Predictive coding = RL + SL + Bayes + MPC
Steven Byrnes
6y
7
6Minimization of prediction error as a foundation for human values in AI alignment
Gordon Seidoh Worley
6y
0
9Towards an Intentional Research Agenda
romeostevensit
6y
5
13AXRP Episode 32 - Understanding Agency with Jan Kulveit
DanielFilan
1y
0
6Refinement of Active Inference agency ontology
Roman Leventov
2y
0
47Conditioning Predictive Models: Large language models as predictors
evhub, Adam Jermyn, Johannes Treutlein, Rubi J. Hudson, kcwoolverton
3y
3
37Conditioning Predictive Models: Outer alignment via careful conditioning
evhub, Adam Jermyn, Johannes Treutlein, Rubi J. Hudson, kcwoolverton
3y
6
10A reply to Byrnes on the Free Energy Principle
Roman Leventov
3y
0
Add Posts