Ryo Kuroiwa / 黒岩稜

Assistant Professor, National Institute of Informatics

<kuroiwa [at] nii [dot] ac [dot] jp>

Research Interest

Combinatorial Optimization, Operations Research, Heuristic Search, Artificial Intelligence, AI Planning

Publications

Journals

• Kuroiwa R. and Beck, J.C. 2024. Domain-Independent Dynamic Programming. Submitted to Artificial Intelligence. paper code models
• Kuroiwa R., Shleyfman A., Piacentini C., Castro M.P., and Beck, J.C. 2022. The LM-Cut Heuristic Family for Optimal Numeric Planning with Simple Conditions. Journal of Artificial Intelligence Research, 75, 1477-1548. paper code

Refereed Conferences

• Kuroiwa R. and Beck, J.C. 2024. Parallel Beam Search Algorithms for Domain-Independent Dynamic Programming. Proc. the 38th Annual AAAI Conference on Artificial Intelligence. (AAAI 2024) paper supplement code slides poster
  • Errata: In Algorithm 8, |L| < k should be added in the while loop condition. This condition is used in the implementation, but I forgot to include it in the pseudo-code.
• Kuroiwa R., Shleyfman A., and Beck, J.C. 2023. Extracting and Exploiting Bounds of Numeric Variables for Optimal Linear Numeric Planning. Proc. the 26th European Conference on Artificial Intelligence. (ECAI 2023) paper supplement code benchmark
• Kuroiwa R. and Beck, J.C. 2023. Large Neighborhood Beam Search for Domain-Independent Dynamic Programming. Proc. the 29th International Conference on Principles and Practice of Constraint Programming. (CP 2023) paper code slides poster
  • Best Poster Award at the Doctoral Program of CP 2023.
• Kuroiwa R. and Beck, J.C. 2023. Solving Domain-Independent Dynamic Programming Problems with Anytime Heuristic Search. Proc. the 33rd International Conference on Automated Planning and Scheduling. (ICAPS 2023) paper supplement code models slides
  • Best Paper Award Runner-up
• Kuroiwa R. and Beck, J.C. 2023. Domain-Independent Dynamic Programming: Generic State Space Search for Combinatorial Optimization. Proc. the 33rd International Conference on Automated Planning and Scheduling. (ICAPS 2023) paper supplement code models and data slides poster
• Shleyfman A., Kuroiwa R., and Beck, J.C. 2023. Symmetry Detection and Breaking in Cost-Optimal Numeric Planning. Proc. the 33rd International Conference on Automated Planning and Scheduling. (ICAPS 2023) paper code benchmark
• Kuroiwa R. and Beck, J.C. 2022. Biased Exploration for Satisficing Heuristic Search. Proc. the 32nd International Conference on Automated Planning and Scheduling. (ICAPS 2022) paper code
• Kuroiwa R., Shleyfman A., and Beck, J.C. 2022. LM-Cut Heuristics for Optimal Linear Numeric Planning. Proc. the 32nd International Conference on Automated Planning and Scheduling. (ICAPS 2022) paper supplement code benchmarks
• Kuroiwa R., Shleyfman A., Piacentini C., Castro M.P., and Beck, J.C. 2021. LM-cut and Operator Counting Heuristics for Optimal Numeric Planning with Simple Conditions. Proc. the 31st International Conference on Automated Planning and Scheduling. (ICAPS 2021) paper supplement code benchmarks
• Kuroiwa R. and Fukunaga A. 2020. Front-to-Front Heuristic Search for Satisficing Classical Planning. Proc. the 29th International Joint Conference on Artificial Intelligence. (IJCAI-PRICAI 2020) paper code
• Kuroiwa R. and Fukunaga A. 2020. Analyzing and Avoiding Pathological Behavior of Parallel Best-First Search. Proc. the 30th International Conference on Automated Planning and Scheduling. (ICAPS 2020) paper code
• Kuroiwa R. and Fukunaga A. 2019. A Case Study on the Importance of Low-Level Algorithmic Details in Domain-Independent Heuristics. Proc the 12th Symposium on Combinatorial Search. (SOCS 2019) paper
• Kuroiwa R. and Fukunaga A. 2019. On the Pathological Search Behavior of Distributed Greedy Best-First Search. Proc. the 29th International Conference on Automated Planning and Scheduling. (ICAPS 2019) paper
• Kuroiwa R. and Fukunaga A. 2018. Batch Random Walk for GPU-Based Classical Planning. Proc. the 28th International Conference on Automated Planning and Scheduling. (ICAPS 2018) paper

Workshops

• Kuroiwa R., Beck, J.C. 2021. A Branch-and-Cut Approach for a Mixed Integer Linear Programming Compilation of Optimal Numeric Planning. Proc. of ICAPS 2021 Workshop on Heuristic Search for Domain-independent Planning (HSDIP) paper
• Yu L., Kuroiwa R., Fukunaga, A. 2020. Learning Search-Space Specific Heuristics Using Neural Network. Proc. of ICAPS 2020 Workshop on Heuristic Search for Domain-independent Planning (HSDIP) paper code

Software

• Domain-Independent Dynamic Programming
  • Generic dynamic programming solver for combinatorial optimization with a Python interface
• NLM-CutPlan
  • First place in all tracks of the International Planning Competition 2023 Numeric Tracks
  • Description: https://ipc2023-numeric.github.io/abstracts/NLM_CutPlan_Abstract.pdf
• Numeric Fast Downward
  • Numeric AI planner for simple and linear numeric planning
• PPlanner
  • Parallel classical planner using GPU, MPI, and multi-thread

Education

• 09/2020-08/24: Ph.D. student, Department of Mechanical and Industrial Engineering, University of Toronto
  • Thesis: Domain-Independent Dynamic Programming
  • Advisor: J. Christopher Beck
• 03/2020: M.S., Department of General Systems Studies, Graduate School of Arts and Sciences, University of Tokyo
  • Thesis: Analyzing and Avoiding Pathological Behavior of Parallel Best-First Search
  • Advisor: Alex Fukunaga
  • Ichiko Memorial Award (Best Thesis Award)
• 03/2018: B.A. in Informatics, Department of Interdisciplinary Sciences, College of Arts and Sciences, University of Tokyo
  • Thesis: A GPU-Based Classical Planner
  • Advisor: Alex Fukunaga

Teaching

• 01/2023 - 04/2023: Teaching Assistant in MIE376: Mathematical Programming (Optimization)
• 04/2018 - 08/2019: Assisted students with electric circuits in laboratory experiments.
• 09/2018 - 11/2018: Tutored Python programming to first-year undergraduate students.

Awards

• 08/2023: Best Poster Award at the Doctoral Program of the 29th International Conference on Principles and Practice of Constraint Programming (CP 2023).
• 07/2023: Best Paper Award Runner-up at the 33rd International Conference on Automated Planning and Scheduling (ICAPS 2023).
• 07/2023: First place in all tracks of the International Planning Competition 2023 Numeric Tracks at the 33rd International Conference on Automated Planning and Scheduling (ICAPS 2023).
• 03/2020 Ichiko Memorial Award (Best Thesis Award) from the University of Tokyo

Grants and Scholarships

• 2018: Division of Multi-Disciplinary Sciences： International Conference Travel Grant (ICAPS 2018)

Community Service

• Reviewer for AAAI 2021-2024
• Reviewer for ECAI 2023,2024
• Reviewer for Constraints
• Reviewer for Mathematical Programming Computation

Work Experience

• 04/2020 - 08/2020: Data scientist at CyberAgent, Inc
• 10/2019 - 11/2019: NLP developer internship at AI Shift, Inc
• 08/2016 - 09/2016: Network engineer internship at CyberAgent, Inc
• 12/2015 - 03/2020: Part-time machine learning developer at Usagee, Inc

Other Activities

• 2020: Committer of Open Banidt Pipeline
• 04/2019 - 03/2020: Leader of the Studnet Steering Committee of ICT Trouble Shooting Contest 2019
• 10/2016 - 03/2019: Member of the Student Steering Committee of ICT Trouble Shooting Contest 7, 8, 9, and 2018