Logical Constraint

Logical constraints, rules that restrict the possible solutions to a problem, are a central focus in improving the reliability and efficiency of various computational systems. Current research emphasizes integrating logical constraints into machine learning models, particularly large language models and neural networks, using techniques like neuro-symbolic integration, variational frameworks, and constraint-aware optimization algorithms. This work aims to enhance model accuracy, consistency, and explainability while addressing challenges like shortcut satisfaction and computational complexity. The impact spans diverse fields, from improving the factual consistency of AI language models to optimizing resource allocation in on-device deep learning and solving complex combinatorial problems more efficiently.

Papers

August 5, 2022

A Model-Oriented Approach for Lifting Symmetries in Answer Set Programming
Alice Tarzariol
Answer Set Programming Inductive Logic Programming Logical Constraint Model Driven Symmetric Model Symmetry Constraint

July 14, 2022

Component twin-width as a parameter for BINARY-CSP and its semiring generalisations
Ambroise Baril, Miguel Couceiro, Victor Lagerkvist
Strong Generalization Many Parameter Constraint Satisfaction Problem Logical Constraint Parameterized Complexity Binary Analysis

May 14, 2022

Efficient lifting of symmetry breaking constraints for complex combinatorial problems
Alice Tarzariol, Martin Gebser, Mark Law, Konstantin Schekotihin
Participation Constraint Answer Set Programming Cyclic Symmetry Automatic Solver Inductive Logic Programming Logical Constraint Hard Combinatorial Symmetric Model Lifting Task

May 1, 2022

Deep Learning with Logical Constraints
Eleonora Giunchiglia, Mihaela Catalina Stoian, Thomas Lukasiewicz
Deep Learning Neural Model Performance Improvement Background Knowledge Logical Constraint Logical Form Safety Critical Application

March 23, 2022

Constrained Clustering and Multiple Kernel Learning without Pairwise Constraint Relaxation
Benedikt Boecking, Vincent Jeanselme, Artur Dubrawski
Clustering Algorithm Distance Metric Logical Constraint Kernel Learning Constrained Clustering Constraint Relaxation Pairwise Constraint

February 10, 2022

P-split formulations: A class of intermediate formulations between big-M and convex hull for disjunctive constraints
Jan Kronqvist, Ruth Misener, Calvin Tsay
Mixed Integer Splitting Algorithm New Formulation Convex Hull Logical Constraint High Level Reasoning

January 20, 2022

Recursive Constraints to Prevent Instability in Constrained Reinforcement Learning
Jaeyoung Lee, Sean Sedwards, Krzysztof Czarnecki
Markov Decision Process Constrained Reinforcement Learning Deterministic Policy Logical Constraint Stable Learning Probabilistic Constraint Constrained Reinforcement Learning Algorithm

December 22, 2021

Lifting Symmetry Breaking Constraints with Inductive Logic Programming
Alice Tarzariol, Martin Gebser, Konstantin Schekotihin
Answer Set Programming Inductive Logic Programming Logical Constraint Symmetry Constraint Combinatorial Search

Logical Constraint

Papers

A Model-Oriented Approach for Lifting Symmetries in Answer Set Programming

Component twin-width as a parameter for BINARY-CSP and its semiring generalisations

Efficient lifting of symmetry breaking constraints for complex combinatorial problems

Deep Learning with Logical Constraints

Constrained Clustering and Multiple Kernel Learning without Pairwise Constraint Relaxation

P-split formulations: A class of intermediate formulations between big-M and convex hull for disjunctive constraints

Recursive Constraints to Prevent Instability in Constrained Reinforcement Learning

Lifting Symmetry Breaking Constraints with Inductive Logic Programming