@inbook{82bed94beed8411bad4c9c4fc7581ff8,
title = "Understanding SATNet: Constraint Learning and Symbol Grounding",
abstract = "The SATNet framework is a neural architecture designed to learn instances of combinatorial problems by learning the set of logical constraints associated with an instance of the maximum satisfiability problem. This turns out to be quite powerful as SATNet is able to learn instances of a wide variety of combinatorial problems (including certain NP-hard problems). SATNet achieves this capability by combining ideas from semidefinite programming with deep learning where the forward pass of the learning process is solving a combinatorial problem. In addition to providing an alternative paradigm for learning logical structure, research on SATNet has led to advances in end-to-end neuro symbolic training and symbol grounding, which are key problems in neuro symbolic research.",
author = "Paulo Shakarian and Chitta Baral and Simari, {Gerardo I.} and Bowen Xi and Lahari Pokala",
note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.",
year = "2023",
doi = "10.1007/978-3-031-39179-8_9",
language = "English (US)",
series = "SpringerBriefs in Computer Science",
publisher = "Springer",
pages = "89--97",
booktitle = "SpringerBriefs in Computer Science",
}