Active Inference: Third International Workshop, IWAI 2022, Grenoble, France, September 19, 2022, Revised Selected Papers

ISBN-13: 9783031287183 / Angielski

Christopher L. Buckley; Daniela Cialfi; Pablo Lanillos

Opis

This volume constitutes the papers of the 3rd International Workshop on Active Inference, IWAI 2022, held in Grenoble, France, in conjunction with ECML/PKDD, on September 19, 2022.

The 25 revised full papers presented in this book were carefully reviewed and selected from 31 submissions.

Spis treści

Preventing Deterioration of Classification Accuracy in Predictive Coding Networks.- Interpreting systems as solving POMDPs: a step towards a formal understanding of agency.- Disentangling Shape and Pose for Object-Centric Deep Active Inference Models.- Object-based Active Inference.- Knitting a Markov blanket is hard when you are out-of-equilibrium: two examples in canonical nonequilibrium models.- Spin glass systems as collective active inference.- Mapping Husserlian phenomenology onto active inference.- The Role of Valence and Meta-awareness in Mirror Self-recognition Using Hierarchical Active Inference.- World model learning from demonstrations with active inference: application to driving behavior.- Active Blockference: cadCAD with Active Inference for cognitive systems modeling.- Active Inference Successor Representations.- Learning Policies for Continuous Control via Transition Models.- Attachment Theory in an Active Inference Framework: How Does Our Inner Model Take Shape?.- Capsule Networks as Generative Models.- Home run: finding your way home by imagining trajectories.- A Novel Model for Novelty: Modeling the Emergence of Innovation from Cumulative Culture.- Active Inference and Psychology of Expectations: A study of formalizing ViolEx.- AIXI, FEP-AI, and integrated world models: Towards a unified understanding of intelligence and consciousness.- Intention Modulation for Multi-Step Tasks in Continuous Time Active Inference.- Learning Generative Models for Active Inference using Tensor Networks.- A Worked Example of the Bayesian Mechanics of Classical Objects.- A message passing perspective on planning under Active Inference.- Efficient search of active inference policy spaces using k-means.- Value Cores for Inner and Outer Alignment: Simulating Personality Formation via Iterated Policy Selection and Preference Learning with Self-World Modeling Active Inference Agent.- Deriving time-averaged active inference from control principles.