Q-HOPE

939 0

HOPE stands for Humanitarian Operations. Q stands for, you guessed it, Quantum.😏

Q-HOPE is quantum and quantum inspired algorithms to solve mathematical models that help humanitarian operations. These include disaster relief after an earthquake, or a hurricane. The particular motivation for our research came from Turkey. You may recall my Distinguish Lecture at Koc University a while back, in 2021 (see Tayurian Calendar). This was introducing quantum integer programming (see Neo-Quantum Organon, Book 1: QuIP) to them. (Book 2 is Chip-based Ising Computing Machine, see Pocket Quantum Computers, and The Physics of GLIMPSEBook 3 is Quantum Queuing.)

One of the important practical problems in disaster relief is:

What does one do immediately after an earthquake? How do the first responders get to the places that require the most immediate and urgent help? The first responders can be ambulances, they can be fire engines, they can be construction equipment. Some of these may already be in the city while others may have to be brought to the city. The people who are able to travel themselves may want to leave the city and go to areas where there are shelters. Which roads are to be prioritized and allocated to first responders so that they can get to the needy quickly, but not block the people who are trying to leave the city at the same time?

This is a combinatorial problem with an imbedded selfish routing problem that is very hard to solve by classical methods. Quantum Inspired Methods provides an alternative approach and this is what we at Tepper (Ananth Tenneti and Anthony Karahalios) are exploring in this research effort with Sibel Salman, Baris Yildiz, Hamed Vosoughian and Amirreza Pashapour (at Koc).

(You know that I love selfish routing! See the Management Science paper on  OrganJet.)

Cancer Genomics (see Mathematics of Cancer Genomics: A ridiculously Short Introduction). Image Classification (see 2021 Tayur Prize). Hedge Fund Portfolio Optimization (see KaRmA).  And now, First Responder problem.  These are some of the applications of quantum and quantum inspired computing that we are tackling at the Tepper Quantum Group. See top of the post for a summary – cannot believe that it was in the Summer of 2017 (see Having Fun with Grobner Basis and Algebraic Geometry) that I decided to dive into Quantum – of some of the collaborators and contributions in  Q-AHA (Quantum: Algorithms, Hardware, Applications).

A common feature of these disparate applications is that they all can be considered as

NOBCO: Non-linear Objective Binary Constrained Optimization problems.

A general purpose Hybrid Quantum Classical algorithm that we have developed is  GAMA (Graver Augmented Multi-Seed Algorithm) that seems to be particularly well-suited for these complex problems (where traditional classical methods have had limited success) by separating the objective function from the constraints, finding feasible solutions (why we are solving an Ising model), and then using Graver Basis that are obtained from the Kernel of the Constraint Matrix as a Test Set. GAMA is an inherently massively parallel algorithm that is able to find good solutions relatively quickly.

To learn more about Q-HOPE, please join us on September 9th:

COPIOR-HORAF CONFERENCE ON “OPERATIONAL RESEARCH TOOLS FOR THE AFTERMATH OF A DISASTER: THE MAJOR EARTHQUAKE IN TURKEY-SYRIA

A date for your calendar: Sat 9th of Sep 2023 (12.00 – 18.00) in Durham, UK/Hybrid.

My dear friends, my dear OR/MS academics, practitioners, and pracademics

Following an inspiring discussion led by Sridhar (Prof Tayur, CC-ed and thanked here) on that OR can help us decide what to do first after a major disaster, and further discussion with HORAF, under my hat as Chair of COPIOR (Committee of Professors in Or in UK www.copior.org ) and representing the most senior level of the OR community in UK, active bystanders and gatekeepers of the discipline, we decided to put some pro bono time and run a daily conference in September for a very good cause: get some monies for the relief of those affected by the earthquake in Turkey-Syria and at the same illustrate the benefits o f OR, in hard and real times, the ‘science of better’ applied when needed.

So we organise a mini conference on the 9th of Sep 2023 with Hybrid format, f2f in Durham UK, and online for those too far away to attend.

Durham university Business School (DUBS) will sponsor the venue (in the Business School in Durham, Mill Hill Lane Durham University Business School), coffees and delegate packs (where the departments of Finance and Marketing and Management will contribute through the research accounts of Julian, Neil and myself). There will be a dedicated website also hosted and designed by DUBS as will be a professional CFP.

The best paper also will be invited for publication to the OUP IMA Journal of Management Mathematics | Oxford Academic (oup.com)

Best,

Kostas

Since many of us are likely meeting up in Montreal soon for MSOM – and I cannot believe it has been five years since our Handbook of Healthcare Analytics was published – Tinglong and I are inviting the contributors (and their friends) to a reception on Saturday 6/24 at Alice Bar (recall that we did one in 2017 MSOM prior to the publication). Let me also take the opportunity to highlight this at the conference (in addition, of course, to the two talks being given by Savannah Tang, on Split Liver Transplantation, see To Split, or Not to Split):

Dear Finalists:

Congratulations again to all of you and your co-authors for being selected by the Academic Committee as Finalists of the M&SOM Practice-Based Research Competition, and thank you for confirming your presentations in person on Sunday, June 25 at the upcoming MSOM Annual Meeting in Montreal. We are excited to announce that the judges serving on the Industry Judge Panel this year will include Srinivas Bollapragada (GE), Mustafa Sir (Amazon) and Sridhar Tayur (CMU, SmartOps, OrganJet). While some of you may have seen the official announcement through the INFORMS and MSOM mailing lists, I also wanted to directly confirm the schedule of your presentations before that panel:

Sunday, June 25 (all times EDT)

Session A

8:30 am – 9:00 am: Leveraging Consensus Effect to Optimize Ranking in Online Discussion Boards by Gur, Allon and Carlstein (presenter Gad Allon)

9:00 am – 9:30 am: Pooling and Boosting for Demand Prediction in Retail: A Transfer Learning Approach by Liu, Lei, Qi, Geng, Zhang, Hu and Shen (presenter Sheng Liu)

9:30 am – 10:00 am: Got (Optimal) Milk? Pooling Donations in Human Milk Banks with Machine Learning and Optimization by Chan, Mahmood, O’Connor, Stone, Unger, Wong and Zhu (presenter Timothy Chan)

Session B

10:30 am – 11:00 am: Decarbonizing OCP by Digalakis, Bertsimas and Cory-Wright (presenter Ryan Cory-Wright)

11:00 am – 11:30 am: Patient Sensitivity to Emergency Department Waiting Time Announcements by Park, Ouyang, Wang, Savin, Leung and Rainer (presenter Eric Park)

11:30 am – 12:00 pm: Improving Farmers’ Income on Online Agri-platforms: Evidence from the Field by Singhvi, Levi, Rajan and Zheng (presenter Yanchong (Karen) Zheng)

Best regards,

Turgay Ayer and Jérémie Gallien

Co-Chairs, M&SOM Practice-Based Research Competition

Leave a Reply