Видео

Thanks for the great material! What text do the definitions come from?

Moore Elizabeth Johnson Carol Gonzalez Ruth

Interesting, thanks!

What is the benchmark instance gaia100m appearing 4:12 to 4:42 in the video?

@oldironchops 7:12 Client who's trying to improve inventory management says that his suppliers don't deliver on time. And the O.R. analyst thinks that's not relevant? Sounds like a very deterministic mindset. Perhaps the O.R. analyst would benefit from greater stochastic awareness when doing optimization modeling.

Very interesting, thank you.

Just awesome!

Thorsten says that the creature at 7:30 is a mammal, which has some very disturbing biological implications.

There is an extremely annoying noise at 0:43 - 0:44. I had to throw away the headphones to escape the torture. Is it possible to fix it?

1) At 28:29 the slide lists "Huge Scale LP problems" under Simplex method, but that item is not discussed. Please clarify, whether that is a qualifier to "Very sparse LP problems"? Because I thought that, other than restart (as mentioned in next item), as problem gets huge, the advantage shifts in favor of barrier. 2) Can you address PDHG for LP on GPU applied to solving in parallel a large number of very small (perhaps single digit number of variables and constraints) dense continuous LPs all having identical size and structure, but different input data (some of the LPs might be feasible, and others infeasible). Can GPUs be used effectively to do that? If not, are there first order methods which could be exploited on GPUs?

Thank you my GOAT

Why do we ignore π₀ in the objective function of the DW pricing problem? If the minimum reduced cost is of an extreme ray, then there might be an extreme point with smaller reduced cost but we cannot determine that because we compute it without π₀? I do not know what I am missing?

what a time to be alive! Please continue your work integer programming Jesus

Adding for completeness and the aid of other novices: MIP :: "Mixed Integer Programming".

Is Primal-Dual Hybrid Gradient amenable to large speedup if implemented on GPU?

I can't believe this is content that exists now. Thanks! 🎉

Why is your herrSolver is better than Gurobi?

Thank you! Really love the idea of having shorts of this kind and would encourage more!

Thank you!

Thanks for a valuable tutorial. Ive been finding this kind of concrete implementation of decomposition methods (which are rare compared to theoretical tutorials 🥹)

On the slide entitled "Nonlinearity Brings New Challenges" you state that continuous relaxation of nonconvex problems no longer give a lower bound. Surely this is only true for local solutions? For global solutions because the relaxation is a strict superset, a global solution would still be a valid bound, no?

Great work Suresh

Great lecture. Thanks a lot!

Professor Lubbecke gives an excellent presentation. It is also a great resource to have softwares like SCIP and gcg !!!! I would like to know which decomposition techniques other than Dantzig Wolfe and Benders have been investigated for the exact solution of MIPs? Could you recommend me some bibliography on this subject? Thank you very much, greetings from the city of Medellin, Colombia.

brilliant lecture!

I was looking for an example with Dantzig Wolfe decomposition with Big M method to solve infeasibility. If someone has any example file or solver code that will be great. thank you.

Robotized measurement ---> ruclips.net/video/dEW5YO7l5TQ/видео.html For more details see ---> www.nature.com/articles/s43246-022-00235-5

Incredible lecture! Thanks!

Thank you!

Awesome video by the way!! Very clear!

Awesome stuff with lovely explanations. Many thanks for sharing! :)

Does Nair mean that there leaves much small room to introduce the ML learning techniques to solve the MIPs?

The person who gave this talk was Zixuan Cao, not Zhewei Huang (me). Sorry we didn't make it clear.

Excellent presentation! thank you!

There is a mistake in the dual of subproblem (7:06), the constraint sign should be <=, not >=.

there is something wrong with the maxChange constraint

Thanks for the brief overview! As you mentioned that there are some high-level APIs to quickly model scheduling or VRP-like problems, I was wondering if these API's can be used to model a problem that contains both routing and scheduling. An example of such can be home service assignment problem where we need to optimize the total cost to hire servers that serve multiple customers and also schedule their routes such that the customers are served in their preferred time duration. Your thoughts on this would be highly appreciated!

awesome！

Adrian, it's absolutely amazing presentation

Thanks for sharing. It helps me a lot.

Thanks! It helps a lot.

Thnaks for this class!

I lack all the math foundations i think to understand this. Any advice how to develop an understanding? where to start...?

Top!

The 15% bit is gold haha

❤OR

I'm wondering why the SAVINGS option doesn't work for the capacity constraints python script. I use PATH_CHEAPEST_ARC instead and it works also. But back to the point, does anyone knows why SAVINGS is not working? I always get the value 2: ROUTING_FAIL: No solution found to the problem

I have to admit Im not a programmer so Im quite laic but I was always wondering why I need these programming tools if I can write my problem by an algebraic modeling language for example AMPL and just "give" my model to a solver. What are the advantages of formulate my problem in C++ or any other programming language?

Beatiful

I've presented the random permutation approach to the ZIB team on the June 3., 2011 as this seemed to be the most influential factor in my thesis and I've published about it later, see www.sciencedirect.com/science/article/abs/pii/S0305054813001020 resp. num.math.uni-goettingen.de/preprints/files/2013-3.pdf (preprint). Love the expression "poor man's approach", I think it catches the spirit!