Distributional Active Inference

• URL: http://arxiv.org/abs/2601.20985v1
• Date: Wed, 28 Jan 2026 19:36:33 GMT
• Title: Distributional Active Inference
• Authors: Abdullah Akgül, Gulcin Baykal, Manuel Haußmann, Mustafa Mert Çelikok, Melih Kandemir,
• Abstract summary: We present a formal abstraction of reinforcement learning algorithms that spans model-based, distributional, and model-free approaches.<n>This abstraction seamlessly integrates active inference into the distributional reinforcement learning framework, making its performance advantages accessible without transition dynamics modeling.
• Score: 11.60734837821471
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Optimal control of complex environments with robotic systems faces two complementary and intertwined challenges: efficient organization of sensory state information and far-sighted action planning. Because the reinforcement learning framework addresses only the latter, it tends to deliver sample-inefficient solutions. Active inference is the state-of-the-art process theory that explains how biological brains handle this dual problem. However, its applications to artificial intelligence have thus far been limited to extensions of existing model-based approaches. We present a formal abstraction of reinforcement learning algorithms that spans model-based, distributional, and model-free approaches. This abstraction seamlessly integrates active inference into the distributional reinforcement learning framework, making its performance advantages accessible without transition dynamics modeling.

Related papers

• Feature-Based vs. GAN-Based Learning from Demonstrations: When and Why [50.191655141020505]
  This survey provides a comparative analysis of feature-based and GAN-based approaches to learning from demonstrations.<n>We argue that the dichotomy between feature-based and GAN-based methods is increasingly nuanced.
  arXiv  Detail & Related papers  (2025-07-08T11:45:51Z)
• Autonomous Option Invention for Continual Hierarchical Reinforcement Learning and Planning [21.737035951695887]
  This paper presents a novel approach for inventing, representing, and utilizing options.<n>Our approach addresses problems characterized by long horizons, sparse rewards, and unknown transition and reward functions.<n>Our main contributions are approaches for continually learning transferable, generalizable options with symbolic representations.
  arXiv  Detail & Related papers  (2024-12-20T23:04:52Z)
• Demonstrating the Continual Learning Capabilities and Practical Application of Discrete-Time Active Inference [0.0]
  Active inference is a mathematical framework for understanding how agents interact with their environments. In this paper, we present a continual learning framework for agents operating in discrete time environments. We demonstrate the agent's ability to relearn and refine its models efficiently, making it suitable for complex domains like finance and healthcare.
  arXiv  Detail & Related papers  (2024-09-30T21:18:46Z)
• Learning in Hybrid Active Inference Models [0.8749675983608172]
  We present a novel hierarchical hybrid active inference agent in which a high-level discrete active inference planner sits above a low-level continuous active inference controller. We make use of recent work in recurrent switching linear dynamical systems which implement end-to-end learning of meaningful discrete representations. We apply our model to the sparse Continuous Mountain Car task, demonstrating fast system identification via enhanced exploration and successful planning.
  arXiv  Detail & Related papers  (2024-09-02T08:41:45Z)
• Explainable Deep Learning Framework for Human Activity Recognition [3.9146761527401424]
  We propose a model-agnostic framework that enhances interpretability and efficacy of HAR models. By implementing competitive data augmentation, our framework provides intuitive and accessible explanations of model decisions.
  arXiv  Detail & Related papers  (2024-08-21T11:59:55Z)
• Distributionally Robust Model-based Reinforcement Learning with Large State Spaces [55.14361269378122]
  Three major challenges in reinforcement learning are the complex dynamical systems with large state spaces, the costly data acquisition processes, and the deviation of real-world dynamics from the training environment deployment. We study distributionally robust Markov decision processes with continuous state spaces under the widely used Kullback-Leibler, chi-square, and total variation uncertainty sets. We propose a model-based approach that utilizes Gaussian Processes and the maximum variance reduction algorithm to efficiently learn multi-output nominal transition dynamics.
  arXiv  Detail & Related papers  (2023-09-05T13:42:11Z)
• Safe Multi-agent Learning via Trapping Regions [89.24858306636816]
  We apply the concept of trapping regions, known from qualitative theory of dynamical systems, to create safety sets in the joint strategy space for decentralized learning. We propose a binary partitioning algorithm for verification that candidate sets form trapping regions in systems with known learning dynamics, and a sampling algorithm for scenarios where learning dynamics are not known.
  arXiv  Detail & Related papers  (2023-02-27T14:47:52Z)
• Learning to Generate All Feasible Actions [4.333208181196761]
  We introduce action mapping, a novel approach that divides the learning process into two steps: first learn feasibility and subsequently, the objective. This paper focuses on the feasibility part by learning to generate all feasible actions through self-supervised querying of the feasibility model. We demonstrate the agent's proficiency in generating actions across disconnected feasible action sets.
  arXiv  Detail & Related papers  (2023-01-26T23:15:51Z)
• Causal Dynamics Learning for Task-Independent State Abstraction [61.707048209272884]
  We introduce Causal Dynamics Learning for Task-Independent State Abstraction (CDL) CDL learns a theoretically proved causal dynamics model that removes unnecessary dependencies between state variables and the action. A state abstraction can then be derived from the learned dynamics.
  arXiv  Detail & Related papers  (2022-06-27T17:02:53Z)
• Stabilizing Q-learning with Linear Architectures for Provably Efficient Learning [53.17258888552998]
  This work proposes an exploration variant of the basic $Q$-learning protocol with linear function approximation. We show that the performance of the algorithm degrades very gracefully under a novel and more permissive notion of approximation error.
  arXiv  Detail & Related papers  (2022-06-01T23:26:51Z)
• Online reinforcement learning with sparse rewards through an active inference capsule [62.997667081978825]
  This paper introduces an active inference agent which minimizes the novel free energy of the expected future. Our model is capable of solving sparse-reward problems with a very high sample efficiency. We also introduce a novel method for approximating the prior model from the reward function, which simplifies the expression of complex objectives.
  arXiv  Detail & Related papers  (2021-06-04T10:03:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.