Reasoning About Actual Causes in Nondeterministic Domains
Authors: Shakil M. Khan, Yves Lespérance, Maryam Rostamigiv
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | We build on recent preliminary work on actual causation in the nondeterministic situation calculus to formalize more sophisticated forms of reasoning about actual causes in such domains. We investigate the notions of Certainly Causes and Possibly Causes that enable the representation of actual cause for agent actions in these domains. We then show how regression in the situation calculus can be extended to reason about such notions of actual causes. ... In the following, we prove a series of properties, which will be the basis of our proposed extended regression operator. ... The proof of this theorem is similar to that of the regression theorem in the SC (Pirri and Reiter 1999; Reiter 2001), but uses Propositions 3 to 5 for the additional cases. |
| Researcher Affiliation | Academia | 1University of Regina, Regina, SK, Canada 2York University, Toronto, ON, Canada EMAIL, EMAIL, EMAIL |
| Pseudocode | No | The paper defines an "Extended Regression Operator" in "Definition 9" using a series of logical rules and transformations, which details how the regression works. However, this is a formal logical definition and not presented as a structured pseudocode block or algorithm with explicit step-by-step instructions for computational execution. |
| Open Source Code | No | Khan and Soutchanski (2020) reported a Prolog implementation of Batusov and Soutchanski’s (2018) original proposal, which we think can be extended to deal with the nondeterministic case with some effort. |
| Open Datasets | No | Our running example involves a robot navigating between different locations and communicating. ... We will call this NDBAT D1. (This refers to a conceptual example and not an actual dataset.) |
| Dataset Splits | No | The paper uses a conceptual 'running example' to illustrate its theoretical framework, not an actual dataset for experimentation. Therefore, no dataset splits are mentioned or provided. |
| Hardware Specification | No | The paper focuses on theoretical formalizations and proofs related to causation in nondeterministic domains. No experimental evaluation is described, and consequently, no hardware specifications for running experiments are provided. |
| Software Dependencies | No | The paper is theoretical, primarily presenting logical formalisms and proofs. While it mentions a 'Prolog implementation' in the context of prior work, it does not specify any software or libraries with version numbers directly relevant to the methodology or results presented in this paper. |
| Experiment Setup | No | The paper is a theoretical work focusing on formalizing notions of causation and extending regression in the situation calculus. It does not describe any experiments or their setup, thus no hyperparameters or training configurations are provided. |