Our survey on deep learning and polyhedral theory is finally published!

pubsonline.informs.org/doi/10.1287/...

This was a great opportunity to explain what we have been looking at without the constraints of being focused on a specific technical contribution or page limit.

@marouanefl.bsky.social shares great recaps. @thserra.bsky.social also shares similar summaries from the conferences he attends. I think all conferences need something like this.

Disjunctive Benders Decomposition We propose a novel enhancement to Benders Decomposition (BD) that generates valid inequalities for the convex hull of the Benders reformulation, addressing a key limitation of conventional BD-its cuts...

For those interested in Benders decomposition (one of my favorite topics), check this paper by Kaiwen Fang, Inho Sin, and (the amazing) Geunyeong Byeon

arxiv.org/abs/2506.03561

DualSchool: How Reliable are LLMs for Optimization Education? Consider the following task taught in introductory optimization courses which addresses challenges articulated by the community at the intersection of (generative) AI and OR: generate the dual of a li...

If you teach an introduction to optimization, or are a student in such a course, you may wonder how LLMs will change the class. Check this paper on Dual School, which provides some answers. @ai4opt.bsky.social @georgiatechai.bsky.social
arxiv.org/abs/2505.21775

Cool stuff! Kinda curious what they’re using for scheduling—sounds like it doesn't give optimal! Or-tools?

The COIN-OR Foundation Bridging New Frontiers in Open-Source Optimization with NVIDIA A New Chapter: COIN-OR and NVIDIA cuOpt Join Forces for Open-Source Optimization Optimization has always been at the core of computational research, and now, a new frontier is emerging—one where op…

Exciting progress in LP and MIP! Congrats to #COINOR and @tedralphs on this new milestone! 🎉 www.coin-or.org/2025/03/18/t...

I didn’t even expect it to be used this much.

4. tuned - A system tuning service that uses mathematical optimization including some LP approaches in its more advanced profiles.
5. IRQBalance - Uses optimization techniques similar to LP to distribute hardware interrupts across processors.

2. systemd-oomd- The Out-Of-Memory daemon in systemd uses techniques related to LP for memory pressure management, though it's more heuristic-based than pure LP.
3. ulatencyd - A system daemon that uses LP techniques for process management and resource allocation.

I wonder and I ask to claude: ...
1. **Aptitude** - This package manager for Debian-based systems explicitly uses LP for dependency resolution. Its documentation mentions that it formulates package dependency resolution as a linear programming problem.

Everyone’s talking about AI, but it didn’t meet my expectations. In finding infeasibility in a simple MIP model, o1 and 4o gave only generic suggestions. Worse, for a specific constraint, it suggested inf. Clearly, there’s still room to improve in handling such problems. #AI #OPTIMIZATION

Ever wondered why presenting more facts can sometimes *worsen* disagreements, even among rational people? 🤔

It turns out, Bayesian reasoning has some surprising answers - no cognitive biases needed! Let's explore this fascinating paradox quickly ☺️

Your test is insightful in many ways. By the way, it’s important to note that Copilot falls behind some paid options.

Looking for a great place to pursue a Ph.D. at the crossroads of data science, machine learning & mathematical optimization?

Apply to Iowa’s Business Analytics program!

For more information: tippie.uiowa.edu/phd/phd-busi...

If you would like to work with me in MIP+ML: {first}-{last}@uiowa.edu

"to derive a vectorized version of our Frank-Wolfe algorithm that could work on multiple separation problems at the same time, increasing the computational density of the operations performed and availing of massively parallel compute accelerators like GPUs in the process"by arxiv.org/pdf/2305.12197

bsky.app/profile/frap...

Extract of Gurobipy code for a districting model in Python

Academics, please publish your data and code

We implore academics to publish working code and data to accompany their papers.

In this blog, we explore an optimization model where publishing has been done well.

www.solvermax.com/blog/academi...

#orms #Python

What’s New – Gurobi 12.0 - Gurobi Optimization Experience global nonlinear solving, speed enhancements, dynamic distributed tuning, and enterprise features.

When will we see solvers harnessing the power of GPUs? For now 👏👉
www.gurobi.com/whats-new-gu...

bileveloptimization.org by Martin Schmidt

Sunday read 👏👉Last fifty years of integer linear programming: a focus on recent practical advances - Inria - Institut national de recherche en sciences et technologies du numérique
inria.hal.science/hal-04776866/

Seen something similar in an OR book: 'A professor can advise a max n students.' Simple constraints like this are key to grasping modeling logic!

@emilyltucker.bsky.social following post is making a positive effect 👏👏

New starter pack if you're looking for the #orms community on here: go.bsky.app/SvBND16

Also if you have recommendations of people to add, feel free to send them my way.