Interpreting Dynamical Systems as Bayesian Reasoners

• URL: http://arxiv.org/abs/2112.13523v1
• Date: Mon, 27 Dec 2021 05:44:47 GMT
• Title: Interpreting Dynamical Systems as Bayesian Reasoners
• Abstract summary: A central concept in active inference is that the internal states of a physical system parametrise probability measures over states of the external world. Here we begin the development of a general theory that would tell us when it is appropriate to interpret states as representing beliefs in this way.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: A central concept in active inference is that the internal states of a physical system parametrise probability measures over states of the external world. These can be seen as an agent's beliefs, expressed as a Bayesian prior or posterior. Here we begin the development of a general theory that would tell us when it is appropriate to interpret states as representing beliefs in this way. We focus on the case in which a system can be interpreted as performing either Bayesian filtering or Bayesian inference. We provide formal definitions of what it means for such an interpretation to exist, using techniques from category theory.

Related papers

• Radical AI Interpretability [0.0]
  We develop a framework for interpreting AI systems as agents.<n>We draw on the philosophical tradition of radical interpretation and the tools of mechanistic interpretability.
  arXiv  Detail & Related papers  (2026-06-25T01:58:38Z)
• Toward a Graph-Theoretic Model of Belief: Confidence, Credibility, and Structural Coherence [0.0]
  This paper introduces a minimal formalism for belief systems as directed, weighted graphs.<n>Unlike logical and argumentation-based frameworks, it supports fine-grained structural representation without committing to binary justification status or deductive closure.<n>Its aim is to provide a foundational substrate for analyzing the internal organization of belief systems.
  arXiv  Detail & Related papers  (2025-08-05T14:03:23Z)
• Counterfactual Realizability [52.85109506684737]
  We introduce a formal definition of realizability, the ability to draw samples from a distribution, and then develop a complete algorithm to determine whether an arbitrary counterfactual distribution is realizable. We illustrate the implications of this new framework for counterfactual data collection using motivating examples from causal fairness and causal reinforcement learning.
  arXiv  Detail & Related papers  (2025-03-14T20:54:27Z)
• A Note on an Inferentialist Approach to Resource Semantics [48.65926948745294]
  'Inferentialism' is the view that meaning is given in terms of inferential behaviour. This paper shows how 'inferentialism' enables a versatile and expressive framework for resource semantics.
  arXiv  Detail & Related papers  (2024-05-10T14:13:21Z)
• Estimation of Counterfactual Interventions under Uncertainties [10.674015311238696]
  "What should I have done differently to get the loan approved?" "What should I have done differently to get the loan approved?" "What should I have done differently to get the loan approved?"
  arXiv  Detail & Related papers  (2023-09-15T11:41:23Z)
• A Geometric Notion of Causal Probing [85.49839090913515]
  The linear subspace hypothesis states that, in a language model's representation space, all information about a concept such as verbal number is encoded in a linear subspace. We give a set of intrinsic criteria which characterize an ideal linear concept subspace. We find that, for at least one concept across two languages models, the concept subspace can be used to manipulate the concept value of the generated word with precision.
  arXiv  Detail & Related papers  (2023-07-27T17:57:57Z)
• Logic meets Wigner's Friend (and their Friends) [49.1574468325115]
  We take a fresh look at Wigner's Friend thought-experiment and some of its more recent variants and extensions. We discuss various solutions proposed in the literature, focusing on a few questions.
  arXiv  Detail & Related papers  (2023-07-04T13:31:56Z)
• Concept Activation Regions: A Generalized Framework For Concept-Based Explanations [95.94432031144716]
  Existing methods assume that the examples illustrating a concept are mapped in a fixed direction of the deep neural network's latent space. In this work, we propose allowing concept examples to be scattered across different clusters in the DNN's latent space. This concept activation region (CAR) formalism yields global concept-based explanations and local concept-based feature importance.
  arXiv  Detail & Related papers  (2022-09-22T17:59:03Z)
• Interpreting systems as solving POMDPs: a step towards a formal understanding of agency [0.0]
  We propose a notion of interpretation not just in terms of beliefs but in terms of goals and actions. An agent is then a system together with an interpretation of this system as a POMDP solution.
  arXiv  Detail & Related papers  (2022-09-04T13:40:33Z)
• Quantum state inference from coarse-grained descriptions: analysis and an application to quantum thermodynamics [101.18253437732933]
  We compare the Maximum Entropy Principle method, with the recently proposed Average Assignment Map method. Despite the fact that the assigned descriptions respect the measured constraints, the descriptions differ in scenarios that go beyond the traditional system-environment structure.
  arXiv  Detail & Related papers  (2022-05-16T19:42:24Z)
• On Nested Justification Systems (full version) [13.76408511310322]
  Nested justification systems, first introduced by Denecker et al., allow for the composition of justification systems. We show in this paper that the original semantics for nested justification systems lead to the loss of information relevant for explanations. Furthermore, we show how nested justification systems allow representing fixpoint definitions.
  arXiv  Detail & Related papers  (2022-05-09T20:23:22Z)
• Observers of quantum systems cannot agree to disagree [55.41644538483948]
  We ask whether agreement between observers can serve as a physical principle that must hold for any theory of the world. We construct examples of (postquantum) no-signaling boxes where observers can agree to disagree.
  arXiv  Detail & Related papers  (2021-02-17T19:00:04Z)
• Are Interpretations Fairly Evaluated? A Definition Driven Pipeline for Post-Hoc Interpretability [54.85658598523915]
  We propose to have a concrete definition of interpretation before we could evaluate faithfulness of an interpretation. We find that although interpretation methods perform differently under a certain evaluation metric, such a difference may not result from interpretation quality or faithfulness.
  arXiv  Detail & Related papers  (2020-09-16T06:38:03Z)
• Belief functions induced by random fuzzy sets: A general framework for representing uncertain and fuzzy evidence [6.713564212269253]
  We generalize Zadeh's notion of "evidence of the second kind" We show that it makes it possible to reconcile the possibilistic interpretation of likelihood with Bayesian inference.
  arXiv  Detail & Related papers  (2020-04-24T10:14:54Z)
• A Unified Scientific Basis for Inference [0.0]
  It is shown that a natural extension of this discussion gives a conceptual basis from which essential parts of the formalism of quantum mechanics can be derived. The questions around Bell's inequality are approached by using the conditionality principle for each observer.
  arXiv  Detail & Related papers  (2012-06-22T07:34:08Z)

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.