Heuristics for Combinatorial Optimization via Value-based Reinforcement Learning: A Unified Framework and Analysis
- Orit Davidovich
- Shimrit Shtern
- et al.
- 2025
- arXiv
Orit Davidovich
Title
Applied Math Research Scientist
Bio
Hi, I’m Orit Davidovich. I work at IBM as an applied math research scientist. I spend my days thinking about data-driven optimzation, while experimenting in hands-on ways. I often spend a considerable amount of time thinking about how best to deploy and distribute stuff on a cluster. You can reach me at orit.davidovich at ibm.com
Publications
Going Topological in Multi-risk Extended Markov Ratio Decision Processes
- 2025
- ICAPS 2025
Finding Probably Approximate Optimal Solutions by Training to Estimate the Optimal Values of Subproblems
- Nimrod Megiddo
- Segev Wasserkrug
- et al.
- 2025
- arXiv
A Rigorous Risk-aware Linear Approach to Extended Markov Ratio Decision Processes with Embedded Learning
- 2023
- IJCAI 2023
Optimization with Constraint Learning
- S. Ilker Birbil
- Donato Maragno
- et al.
- 2023
- AAAI 2023
The Good, the Bad, and the Outliers: A Testing Framework for Decision Optimization Model Learning
- Orit Davidovich
- Gheorghe-Teodor Bercea
- et al.
- 2022
- KDD 2022
Addressing Solution Quality in Data-generated Optimization Models
- 2022
- AAAI 2022
Data and Knowledge Driven Optimization Model Generation for Flow Based Optimization Problems
- Dharmashankar Subramanian
- Segev Wasserkrug
- et al.
- 2021
- INFORMS 2021
The Combinatorial Generation of Objective Targets and Constraints for Large-scale Testing of Optimization Routines
- 2021
- FWCG 2021
Ensuring the Quality of Optimization Solutions in Data Generated Optimization Models
- 2021
- IJCAI 2021
Blog posts
DOFramework: A testing framework for decision optimization model learners
Technical noteOrit Davidovich
Top collaborators
SW
Segev Wasserkrug
STSM, Core Math team, Optimization
PR
Parikshit Ram
Research Staff Member
DP
Dzung Phan
Research Staff Member
PM
Pavankumar Murali
Senior Research Scientist & Master Inventor