Learning Abduction Using Partial Observability

Authors: Brendan Juba, Zongyi Li, Evan Miller

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this work, we consider a learning to reason (Khardon and Roth 1997) or PAC-learning (Valiant 1984; 2000) formulation of the combined task of learning to abduce, introduced by Juba (2016).
Researcher Affiliation Academia Brendan Juba, Zongyi Li, Evan Miller Dept. of Computer Science and Engineering Washington University in St. Louis {bjuba, zli, evan.a.miller} @wustl.edu
Pseudocode Yes Algorithm 1: Decide PAC
Open Source Code No The paper does not contain any explicit statements about releasing source code or provide links to a code repository for the methodology described.
Open Datasets No This paper is theoretical and focuses on a PAC-learning framework with conceptual 'examples drawn from the prior distribution' rather than using a specific, publicly available dataset for empirical evaluation.
Dataset Splits No As a theoretical paper, no empirical experiments with validation dataset splits are described.
Hardware Specification No The paper is theoretical and does not describe any experiments that would require specific hardware, thus no hardware specifications are provided.
Software Dependencies No The paper does not provide specific software dependencies with version numbers for replication.
Experiment Setup No The paper is theoretical and presents algorithms and proofs, but does not describe any empirical experimental setups or hyperparameter details.