Trustworthy Planning and Scheduling with Learning and Explanations

Project Information

TUPLES

Grant agreement ID: 101070149

DOI 10.3030/101070149

EC signature date 12 June 2022

Start date 1 October 2022

End date 30 September 2025

Funded under

• Digital, Industry and Space

Total cost € 4 200 886,25

EU contribution € 4 200 886,25

Coordinated by

UNIVERSITE DE TOULOUSE
France

Deliverables

Other (6)

GitHub/TUPLES lab - simulators & planning domains

Use case simulators and planning domains hosted on GitHub/TUPLES lab (M18, Software).

Fram. learning with robustness guarantees

Framework for policy learning with decision-robustness guarantees (M18, Software).

Expl. ML models and systems of constraints

Explanations for ML models and systems of constraints (M12, Report+Software).

Fram. for robustness verification of policies

Framework for robustness verification of policy-decisions (M18, Software).

Expl. policies in non-learning based P&S

Explanations for action policies in non-learning based scheduling and planning (M18, Report+Software).

Methods for integration of LRNNs into P&S

Methods for integration of LRNNs into planning and scheduling domains (M18, software, report)

Documents, reports (2)

Integrated online/offline (re-)scheduling

Integrated offline/online scheduling and rescheduling solutions (M18, Report).

Robust integration of learning and solving

Robust integrated learning and solving with predicted costs (M18, Report).

Publications

Conference proceedings (26)

Adversarial Example Detection in Deployed Tree Ensembles

Author(s): Laurens Devos, Lorenzo Perini, Wannes Meert, Jesse Davis
Published in: Machine Learning and Knowledge Discovery in Databases; 2023, 2023
Publisher: Springer
DOI: 10.48550/arXiv.2206.13083

Action Policy Testing with Heuristic-Based Bias Functions

Author(s): Xandra Schuler, Jan Eisenhut, Daniel Höller, Daniel Fišer and Jörg Hoffmann
Published in: Proceedings of the ICAPS Workshop on Reliable Data-Driven Planning and Scheduling (RDDPS), 2023
Publisher: none

Safety Verification of Tree-Ensemble Policies via Predicate Abstraction

Author(s): Chaahat Jain, Lorenzo Cascioli, Laurens Devos, Marcel Vinzent, Marcel Steinmetz, Jesse Davis, Jörg Hoffmann
Published in: Proceedings of the ICAPS Workshop on Reliable Data-Driven Planning and Scheduling (RDDPS'24), 2024
Publisher: None

Neural Policy Safety Verification via Predicate Abstraction: CEGAR

Author(s): Marcel Vinzent, Siddhant Sharma, Jöerg Hoffmann
Published in: Proceedings of the 37th AAAI Conference on Artificial Intelligence (AAAI), 2023, ISSN 2374-3468
Publisher: AAAI Press
DOI: 10.1609/aaai.v37i12.26772

Simplifying Step-wise Explanation Sequences

Author(s): Bleukx, Ignace; Devriendt, Jo; Gamba, Emilio; Bogaerts, Bart; Guns, Tias
Published in: Leibniz International Proceedings in Informatics; 2023, 2023, ISSN 1868-8969
Publisher: Leibniz International Proceedings in Informatics; 2023

Guided Bottom-Up Interactive Constraint Acquisition

Author(s): Dimosthenis Tsouros, Senne Berden, Tias Guns
Published in: Leibniz International Proceedings in Informatics; 2023, 2023, ISSN 1868-8969
Publisher: Schloss Dagstuhl -- Leibniz-Zentrum fuer Informatik
DOI: 10.48550/arXiv.2307.06126

Optimize Planning Heuristics to Rank, not to Estimate Cost-to-Goal

Author(s): Leah Chrestien, Stefan Edelkamp, Antonin Komenda, Tomas Pevny
Published in: Proceedings of the 37th Annual Conference on Neural Information Processing Systems (NeurIPS'23), 2023
Publisher: none

Blossom: an Anytime Algorithm for Computing Optimal Decision Trees

Author(s): Emir Demirović, Emmanuel Hebrard, Louis Jean
Published in: Proceedings of the 40th International Conference on Machine Learning (ICML), 2023
Publisher: PMLR

Computational Asymmetries in Robust Classification

Author(s): Samuele Marro, Michele Lombardi.
Published in: 40th International Conference on Machine Learning (ICML 2023), 2023
Publisher: ICML
DOI: 10.48550/arXiv.2306.14326

Policy-Specific Abstraction Predicate Selection in Neural Policy Safety Verification

Author(s): Marcel Vinzent, Min Wu, Haoze Wu, Jörg Hoffmann
Published in: Proceedings of the ICAPS Workshop on Reliable Data-Driven Planning and Scheduling (RDDPS), 2023
Publisher: none

Learning to Learn in Interactive Constraint Acquisition

Author(s): Dimosthenis Tsouros, Senne Berden, Tias Guns
Published in: Proceedings of the AAAI Conference on Artificial Intelligence, Issue 38, 2024, ISSN 2159-5399
Publisher: AAAI Press, Washington, DC, USA
DOI: 10.1609/aaai.v38i8.28655

Learning domain-independent heuristics for grounded and lifted planning

Author(s): Dillon Z. Chen, Sylvie Thiébaux, Felipe Trevizan
Published in: Proceedings of the 38th AAAI Conference on Artificial Intelligence (AAAI-24), 2024, ISSN 2159-5399
Publisher: AAAI Press
DOI: 10.1609/aaai.v38i18.29986

Guiding GBFS through Learned Pairwise Rankings

Author(s): Mingyu Hao, Felipe Trevizan, Sylvie Thiebaux, Patrick Ferber and Jörg Hoffmann
Published in: Proceedings of the 33rd International Joint Conference on Artificial Intelligence (IJCAI'24), 2024
Publisher: IJCAI Organization

Towards Feasible Higher-Dimensional Potential Heuristics

Author(s): Daniel Fiser, Marcel Steinmetz
Published in: Proceedings of the 34th International Conference on Automated Planning and Scheduling (ICAPS-24), 2024
Publisher: AAAI Press

Automatic Metamorphic Test Oracles for Action-Policy Testing

Author(s): Jan Eisenhut, Álvaro Torralba, Maria Christakis, Jörg Hoffmann
Published in: Proceedings of the 33rd International Conference on Automated Planning and Scheduling (ICAPS), 2023
Publisher: AAAI Press
DOI: 10.1609/icaps.v33i1.27185

Novelty Heuristics, Multi-Queue Search, and Portfolios for Numeric Planning

Author(s): Dillon Z. Chen, Sylvie Thiébaux
Published in: Proceedings of the 17th International Symposium on Combinatorial Search (SOCS-24), 2024
Publisher: AAAI Press
DOI: 10.48550/arXiv.2404.05235

Return to Tradition: Learning Reliable Heuristics with Classical Machine Learning

Author(s): Dillon Z. Chen, Felipe Trevizan, Sylvie Thiébaux
Published in: Proceedings of the 34th International Conference on Automated Planning and Scheduling (ICAPS-24), 2024
Publisher: AAAI Press
DOI: 10.48550/arXiv.2403.16508

Assessing the Transparency and Explainability of AI Algorithms in Planning and Scheduling Tools: A Review of the Literature

Author(s): Sofia Morandini, Federico Fraboni, Enzo Balatti, Aranka Hackmann, Hannah Brendel, Gabriele Puzzo, Lucia Volpi, Davide Giusino, Marco De Angelis, Luca Pietrantoni.
Published in: AHFE (2023) International Conference, 2023
Publisher: AHFE International
DOI: 10.54941/ahfe1004068

Explainability Insights to Cellular Simultaneous Recurrent Neural Networks for Classical Planning

Author(s): Michaela Urbanovská, Antonín Komenda
Published in: Proceedings of the 16th International Conference on Agents and Artificial Intelligence (ICAART'24), Issue Volume 3, 2024, ISSN 2184-433X
Publisher: SciTePress
DOI: 10.5220/0012375800003636

Score Function Gradient Estimation to Widen the Applicability of Decision-Focused Learning

Author(s): Mattia Silvestri, Senne Berden, Jayanta Mandi, Ali Mahmutogullari, Maxime Mulamba, Allegra De Filippo, Tias Guns, Michele Lombardi
Published in: Differentiable Almost Everything workshop @ICML 2023, 2023
DOI: 10.48550/arXiv.2307.05213

Robustness Verification of Multi-Class Tree Ensembles

Author(s): Laurens Devos, Lorenzo Cascioli, Jesse Davis
Published in: Proceedings of the AAAI Conference on Artificial Intelligence, Issue 38, 2024, ISSN 2159-5399
Publisher: AAAI Press, Washington, DC, USA
DOI: 10.1609/aaai.v38i19.30093

Learning Generalised Policies for Numeric Planning

Author(s): Ryan Xiao Wang, Sylvie Thiébaux
Published in: Proceedings of the 34th International Conference on Automated Planning and Scheduling (CAPS-24), 2024
Publisher: AAAI Press

On the Computational Complexity of Stackelberg Planning and Meta-Operator Verification

Author(s): Gregor Behnke; Marcel Steinmetz
Published in: 34th International Conference on Automated Planning and Scheduling (ICAPS-24), 2024
Publisher: AAAI Press

Action Policy Explanations in Oversubscription Planning

Author(s): Aleena Siji, Rebecca Eifler, Daniel Fišer, Jörg Hoffmann
Published in: Proceedings of the ICAPS Workshop on Human-Aware and Explainable Planning (HAXP), 2023
Publisher: none

Fast and robust resource - constrained scheduling with graph Neural Networks

Author(s): Teichteil-Koenigsbuch, F., Poveda,g., Gonzales de Garibay Barba, G., Luchterhand, T. and Thiébaux, S.
Published in: Proceedings of the 33rd International Conference on Automated Planning and Scheduling (ICAPS), 2023
DOI: 10.1609/icaps.v33i1.27244

Formal Explanations of Neural Network Policies for Planning

Author(s): Renee Selvey, Alban Grastien, Sylvie Thiébaux
Published in: Proceedings of International Joint Conference on Artificial Intelligence (IJCAI), 2023, 2023
Publisher: IJCAI.org
DOI: 10.24963/ijcai.2023/605

Peer reviewed articles (2)

A Markov Framework for Learning and Reasoning About Strategies in Professional Soccer

Author(s): Maaike Van Roy, Peter Robberechts, Wen Chi Yang, Luc De Raedt, Jesse Davis
Published in: Journal Of Artificial Intelligence Research, Issue 77, 2023, ISSN 1076-9757
Publisher: Morgan Kaufmann Publishers, Inc.
DOI: 10.1613/jair.1.13934

An efficient heuristic for very large-scale vehicle routing problems with simultaneous pickup and delivery

Author(s): Francesco Cavaliere, Luca Accorsi, Demetrio Laganà, Roberto Musmanno, Daniele Vigo
Published in: Transportation Research Part E: Logistics and Transportation Review, Issue Volume 186, June 2024, 2024, ISSN 1878-5794
Publisher: Elsevier
DOI: 10.1016/j.tre.2024.103550