Decision Optimization

Decision Optimization

We’re passionate about the infusion of structured methods, such as mathematical optimization, game theory and reinforcement learning into decision-making processes at scale, thereby using AI to help make better decisions and providing significant benefits to enterprises.

 

We’re passionate about the infusion of structured methods, such as mathematical optimization, game theory and reinforcement learning  into decision making processes at scale, thereby using AI to help make better decisions and providing significant benefits to enterprises.

We invite collaborations in our core research areas:

1.
Scientific methods at the intersection of data science, AI, and optimization intended to scale the use of optimization solutions by reducing by  orders of magnitude the time and skills required to create optimization models.

2.
Advancing the science of multi-party decision making in enterprise scenarios, through theoretical advancements and practical application of game theory, algorithmic game theory, and multi-agent reinforcement learning.

3.
Creation of new optimization algorithms and methods.

 

Collaboration

Scaling optimization usage

Driving the science of end-to-end prediction to optimization pipelines by taking advantage of the fact that the ultimate goal of the prediction pipeline is optimization.

With Bistra Dilkina (University of Southern California) and Andrea Lodi (Jacobs Technion-Cornell Institute).

 

Definitional deficiencies of game theory

Games assume a fixed set of players. There's a need for new solution concepts accounting for players being able to leave the game.

With Miklós Pintér (Budapest University of Technology and Economics) and Eilon Solan (Tel-Aviv University)

Computational tractable multi-party solution concepts

Game theory typically assumes unbounded computational ability. There's a requirement for computationally efficient algorithms for centralized and coordinated multi-party decision making.

With Mathias Staudigl (Maastricht University), Barbara Franci (Barbara Franci) and Matúš Mihalák (Maastricht University).

Dealing with partial knowledge in real world settings

Reasoning with partial knowledge such as incomplete information about other players, partial information revelation, and incentivizing truthful knowledge revelation.

With János Flesch (Maastricht University), Miklós Pintér (Budapest University of Technology and Economics), and Elisheva Shamash (Technion).

Sample Publications

Title Author Year Conference/Journal  
Ensuring the Quality of Optimization Solutions in Data Generated Optimization Models Segev Wasserkrug, Orit Davidovich, Evegeny Shindin, Dharmashankar Subramanian, Parikshit Ram, Pavankumar Murali, Dzung Phan, Nianjun Zhou and Lam M. Nguyen 2021 Data Science Meets Optimisation, IJCAI 2021 Workshop Link
Ranking Data Slices for ML Model Validation: A Shapley Value Approach Eitan Farchi, Ramasuri Narayanam, Lokesh Nagalapatti 2021 2021 IEEE 37th International Conference on Data Engineering (ICDE) Link
Broadly Applicable Targeted Data Sample Omission Attacks Guy Barash, Eitan Farchi, Sarit Kraus, and Onn Shehory 2021 International Workshop on Safety and Security of Machine Learning Link
A Game Theoretic Model for Strategic Coopetition in Business Networks Segev wasserkrug and Eitan Farchi 2021 INFORMS 2021 annual meeting  
A Game Theoretic Model for Strategic Coopetition in Business Networks Segev wasserkrug and Eitan Farchi 2020 Wine 2020 Game Theory for Blockchain Workshop Link
Attention Actor-Critic algorithm for Multi-Agent Constrained Co-operative Reinforcement Learning P. Parnika, Raghuram Bharadwaj Diddigi, Sai Koti Reddy Danda, and Shalabh Bhatnagar 2021 AAMAS 2021 Link
Learner-Independent Data Omission Attacks Guy Barash, Onn Shehory, Sarit Kruas, Eitan Farchi 2020 EDSMLS 2020 Link
IBM Crew Pairing and Rostering Optimization (C-PRO) Technology with MDP for Optimization Flow Orchestration Vladimir Lipets, and Alexander Zadorojniy 2021 Springer Nature Link

Industrial student projects

If you are a student with a relevant background in exact science disciplines or engineering and want to hear more about this and other opportunities, please contact us.

Empirical analysis of learning in multi-party decision making settings

Dr. Segev Wasserkrug , Dr. Eitan Farchi and Dr. Takayuki Osogami, at the Technion

Awards

Regret minimization in supply chain games 2021 Faculty Award

Nicolò Cesa-Bianchi, Professor of Computer Science, Dipartimento di Informatica & Data Science Research Center, Università degli Studi di Milano, Italy

Research on continuous-time linear programming 2020 Faculty Award

Gideon Weiss, Professor Emeritus, University of Haifa

 
 

Talks, workshops
& tutorials

AI4DO Workshop

The AAAI-22 workshop on AI for Decision Optimization (AI4DO) will explore how AI can be used to significantly simplify the creation of efficient production level optimization models, thereby enabling their much wider application and resulting business value.The desired outcome of this workshop is to drive forward research and seed collaborations in this area by bringing together machine learning and decision-making from the lens of both dynamic and static optimization models.

Learn more

Dharmashankar Subramanian & Segev Wasserkrug

Radical simplification for the creation of optimization models

Talk at INFORMS 2021

Alexander Zadorojniy, Takayuki Osogami

A Strongly Polynomial Algorithm For Risk Constrained Problems

Talk at INFORMS 2021

Segev Wasserkrug, Eitan Farchi, Nimrod Megiddo

A Game Theoretic Model For Strategic Coopetition In Business Networks

Talk at INFORMS 2021

More talks

Alexander Zadorojniy & Segev Wasserkrug
MDP Graph-based Intermediate Model for DRL Training
Talk at INFORMS 2020

Dharmashankar Subramanian, Segev Wasserkrug, Pavankumar Murali,
Dzung Phan, Parikshit Ram, Orit Davidovich, Xavier Ceugniet,
Ferenc Katai

Data And Knowledge Driven Optimization Model Generation For Flow Based Optimization Problems
Talk at INFORMS 2021

Ferenc Katai, Evgeny Shindin
Multi-objective Optimization and its Pareto Extension
Talk at INFORMS 2021

Segev Wasserkrug, Alexander Zadorojniy, Dharmashankar Subramanian
Automated Derivation Of MDP And Reinforcement Learning Models From Historical Data
Talk at INFORMS 2020

Let's talk

Want to collaborate? We're always happy to talk. Feel free to get in touch.

 


Decision Optimization Area Focal Point

Working together
with IBM Research - Israel

Our published papers

Our catalog of recent publications authored by IBM researchers, in collaboration with the global research community.


Explore

Open positions and internships

Join some of the best scientists from around the world, at IBM Research - Israel. This is a list of the current open positions, including internships and and student opportunities.

Meet our interns
Explore

Grants &
fellowship

IBM supports students in their research, encourages and invites the Fellowship winners to do an internship in order to be better informed with the the research done by the company.


Explore

EU funded
projects

IBM Research - Israel is involved in over 30 EU-funded consortiums and conducts numerous joint projects with academic and industrial partners around the world.


Explore