I did a bit
to break
the spell
We have to do it another way: it's part of our age-old task to escape the supremacy of business&networking&big-talking (rev 3:8)
From a small implementation detail to a real manifesto on the artificial&injurious gap between theory and real computing.htm
On the tragic side of working for the truth.pdf


Daniel Porumbel

Associate professor (MCF),  Combinatorial Optimization TeamCEDRIC CS Lab, CNAM, Paris
Contact: daniel.porumbel@removeme.cnam.fr, +(33)1 40 27 24 33


Research Context

Expository/submitted papers

Daniel Porumbel, Demystifying the characterizations of SDP matrices in mathematical programming.pdf: I wrote this introduction to SDP (semipositive definite) programming becase I found no other introduction that targets the same public. This work is intended to be accessible to anybody who learned (a lot of) maths in the past but does not work full time in (SDP) matrix theory. For instance, the eigen-decompostion is not a pre-requisite unlike in most other SDP introductions; I even provide two proofs to gain full insight and to learn related properties. All in 100 pages, not 700. The text is self-contained in a strong sense, i.e., everything is proven from scratch, because I can't stand taking things for granted: no secret, no empty jargon, no more formula out of the blue, no longer my(s)thification!
(here is a document in the same spirit on the KKK conditions for interior point linear programming).

New idea: Daniel Porumbel, Semidefinite Programming by Projective Cutting Planes (pdf conf submission)

Amelie Lambert, Daniel Porumbel, A piecewise-quadratic convexification for exactly solving box-constrained quadratic programs (pdf draft) Accepted at World Congress on Combinatorial Optimization 2023

International Journals

[J1] Daniel Porumbel, Projective Cutting-Planes, SIAM Journal on Optimization, 30(1): 1007-1032, 2020 (pdf draft, slides, source code, youtube 20 minutes, youtube 40 minutes, doi link)

[J2] Daniel Porumbel, Ray Projection for Optimizing Polytopes with Prohibitively Many Constraints in Set-Covering Column Generation, Mathematical Programming 155(1): 147-197, 2016 (pdf draft, slides 2013, source code), Springer©, doi link

[J3] Daniel Porumbel, Further experiments and insights on Projective Cutting-Planes pdf, this technical report CEDRIC-19-4550 was initially linked to [J1], but it become Further experiments and insights on Projective Cutting-Planes published by INFORMS Journal of Computing 34:2736-2753 (2022)

[J4] Daniel Porumbel, From the Separation to the Intersection Sub-problem in Benders Decomposition Models with Prohibitively-Many Constraints, Discrete Optimization, 29:148-173, 2018 (pdf draft, source code, slides [fr]),

[J5] Daniel Porumbel, François Clautiaux, Convergent Dual Bounds Using an Aggregation of Set-Covering Constraints for Capacitated Problems, INFORMS Journal of Computing, 29(1):170-184, 2017. (pdf draft, slides, supplement)

[J6] Daniel Porumbel, J-K. Hao, Distance-Guided Local Search, Journal of Heuristics, 2020, 26(5):711-741. (pdf draft)

[J7] D. C. Porumbel, J-K. Hao, P. Kuntz, A Search Space "Cartography" for Guiding Graph Coloring Heuristics. Computers & Operations Research, 37(4):769-778, 2010. (pdf draft), Elsevier©, doi link

[J8] Daniel Porumbel, Prize-Collecting Set Multi-Covering With Submodular Pricing, International Transactions in Operational Research, 25 (4):1221-1239, 2018, (pdf draft), Wiley-Blackwell©

[J9] D. Porumbel, G. Goncalvez, H. Allaoui; T. Hsu; Iterated Local Search and Column Generation to solve Arc-Routing as a permutation set-covering problem, European Journal of Operational Research 256(2):349-367, 2017 pdf draft, slides Roadef 2015 (French OR congress) Elsevier© ,

[J10] D. Porumbel, G. Goncalves, Using Dual Feasible Functions to Construct Fast Lower Bounds for Routing and Location Problems, Discrete Applied Mathematics, 196: 83-99, 2015, (pdf draft), Elsevier©, doi

[J11] Daniel Porumbel. Isomorphism Testing using Polynomial-Time Graph Extensions. Journal of Mathematical Modelling and Algorithms, 10(2):119-143, 2011 (pdf draft, slides [fr], isomorphism program GI-Ext), Springer©, doi

[J12] D. C. Porumbel, J-K. Hao, P. Kuntz. An Efficient Algorithm for Computing the Distance Between Close Partitions. Discrete Applied Mathematics, 159(1):53-59, 2011, (pdf draft), Elsevier©, doi

[J13] D. C. Porumbel, Jin-Kao Hao, Fred Glover, A Simple and Effective Algorithm for the MaxMin Diversity Problem. Annals of Operations Research, 186 (1):275-293, 2011, (pdf draft, source code), Springer©, doi

[J14] D. C. Porumbel, J-K. Hao, P. Kuntz. An Evolutionary Approach with Diversity Guarantee and Well-Informed Grouping Recombination for Graph Coloring, Computers & Operations Research, 37(10):1822-1832, 2010 (pdf draft), Elsevier©, doi

[J15] D. Porumbel, I. Machao, E-G. Talbi, Using an Exact Bi-Objective Decoder in a Memetic Algorithm for Arc-Routing (and Other Decoder-Expressible) Problems European Journal of Operational Research, in press 2023, (pdf draft), Elsevier©, doi

[J16] Nathalie Helal, Frédéric Pichon, D. Porumbel, David Mercier, Éric Lefèvre. The capacitated vehicle routing problem with evidential demands, International Journal of Approximate Reasoning 95:124-151, 2018 (pdf) Elsevier©

[J17] D. C. Porumbel, J-K. Hao, P. Kuntz. Informed Reactive Tabu Search for Graph Coloring. Asia Pacific Journal of Operations Research, 30(4), pp. 1350010, 2013 (pdf draft), World Scientific Publishing©

Conferences and Book Chapters:

[C1] D. Porumbel, J-K. Hao, P. Kuntz, Spacing Memetic Algorithms. In proceedings of GECCO 2011, GA track, 1061-1068. ACM. (pdf draft, slides)

[C2] D. Porumbel, G. Goncalves. Optimizing Set-Covering Models when Column Generation requires Knapsack Based Subproblems with Unbounded Capacities. (conference based on a short one-page submission) The 27th Conference of the European Chapter on Combinatorial Optimization (ECCO Meating), Paris, 2013

[C3] G. Audemard, C. Lecoutre, M. S. Modeliar, G. Goncalves, D. Porumbel Scoring-based Neighborhood Dominance for the Subgraph Isomorphism Problem. The 20th International Conference on Principles and Practice of Constraint Programming (CP 2014)

[C4] Philippe Galinier, Jean-Philippe Hamiez, Jin-Kao Hao, Daniel Porumbel, Recent advances in graph vertex coloring, In I. Zelinka, V. Snasel, A. Abraham (eds.), Handbook of Optimization, Springer© (Intelligent Systems Series)

[C5] D. Porumbel, J-K. Hao, P. Kuntz, Diversity Control and Multi-Parent Recombination for Evolutionary Graph Coloring Algorithms. Evocop 2009 (9th European Conference on Evolutionary Computation in Combinatorial Optimisation, Tübingen, Germany), LNCS 5482: 121-132, 2009. (ps, pdf) Springer©  among the three best paper nominees

[C6] D. Porumbel, J-K. Hao, P. Kuntz, Position-Guided Tabu Search for Graph Coloring. Accepted for the post-proceedings of LION 2009 (Learning and Intelligent OptimizatioN conference, Trento, Italy), LNCS 5851:148-162, 2009. (paper draft, slides) Springer©

[C7] D. Porumbel, J-K. Hao, P. Kuntz, A study of evaluation functions for the graph K-coloring problem. Selected papers from the 8th International Conference on Artificial Evolution (Tours, France), LNCS 4926: 124-135, 2008. (ps, pdf) Springer© 

Expository work, reports, PhD

As far as I know, there is no relevant technical mistake in articles below. If you are interested to (re-)work, extend or (re)-submit, you can contact me.

[E1] D. Porumbel, Revisiting the bijection between planar maps and well labeled trees (pdf). In this work, we provide alternative arguments and correct a slight error of a graph theory article from around 1970s/1980s.

[PhD] D. Porumbel, PhD thesis, English version, French version, dedicated to my grandfather.

More[+] (Seminars, theses)



Software

I like writing software more then delegating. I try to systematically provide the source code for all my new algorithmic work, except for papers without implementation [J8, J12]. You can check my CODE_GUIDELINES used by most recent C++ projects. Exemples:

Projective Cutting Planes for Benders decompositions, Column Generation, a robust programming problem, ...
Replacing the separation sub-problem with the intersection sub-problem to solve a Benders reformulation model (2018)
Distance Guided Local Search for maximum weighted k-cluster or k-clique (2018)
Integer Ray Method for Column Generation for Cutting Stock and Capacitated Arc Routing (2014)
Heuristic for maximum minimum diversity (2011)
Isomorphism tool GI-Ext (2010)

Scientific Awards

I did win some modest science prizes throughout my life. Even since I was a kid I enjoyed math problem-solving contests and I become a International Mathematical Olympiad medalist. Even later, in 2011, I received the PhD prize Simon Régnier awarded by the "Société Francophone de Classification" (French-speaking Classification Society), see these slides. I now believe that science prizes make sense to spark enthusiasm in kids or students. For a grown-up man, thinking about prizes (even up to the Nobel one) is short-sighted or works for the spirit of business. The only prize I want is you, or you and me in the same train of thought. I even thank you for stopping by, whoever you may be.

Nominated for Best Paper Award (3 papers selected among more than 50), EvoCOP 2009

Bronze medal at the 41st International Mathematical Olympiad, a problem-solving competition with contestants from almost 100 countries (ranked 2nd in the qualification rounds of Romania), South Korea, 2000

Diploma for Academic Excellence of Romania's President Emil Constantinescu, 2000



Service for the Scientific Community

I accepted review invitations from the editorial board of different journals since 2017: INFORMS Journal of Computing, Computational Optimization and Applications, European Journal of Operational Research, Operational Research , IEEE Transactions on Cybernetics, Computers and Operations Research, Transportation Research part E, Engineering Optimization, Theoretical Computer Science, Discrete Applied Mathematics, Journal of Computational Science. Before 2017, I provided reviews for journals like: Canadian Mathematical Bulletin, Journal of Artificial Intelligence Research, The Computer Journal, IEEE Transactions on Evolutionary Computation, European Journal of Operations Research, Computers and Operations Research, Evolutionary Computation, Computers & Mathematics with Applications, Information Processing Letters, Expert Systems and Applications, Engineering Applications of Artificial Intelligence, Computing. I was in the program committee of different conferences (Evocop 2012, Evocop 2013, Evocop 2014, Evocop 2019, Evocop 2020, Evocop 2021, Evocop 2022, Evocop 2023, Gecco Ecom 2013, Gecco Ecom 2014, Gecco Ecom 2015, Gecco Ecom 2016, Gecco Ecom 2017, Gecco Ecom 2018, Gecco Ecom 2019, Gecco Ecom 2020, Gecco Ecom 2021, Gecco Ecom 2022, Gecco Ecom 2023, Ecta 2019-2021). I acted as subreviewer for SEA 2023.

I (co-)advised a PhD thesis: Nathalie Helal, An evidential answer for the capacitated vehicle routing problem with uncertain demands, defended in 2017 (main adviser: Eric Lefèvre, second adviser Frédéric Pichon, collaboration with David Mercier). I adivised some master theses along the years: Thomas Ridremont (Robust optimisation of wiring in renewable-energy installations, with Marie-Christine Costa and Cédric Bentz, 2015); Hubert Arnoux (A tool of visualisation techniques for local search methods, with J-K Hao, P. Kuntz, 2012); Junlin Zhu (Mathematical Programming Models for Graph Isomorphism, with Amélie Lambert, 2015); Ben Ibrahima Badji (Meta-heuristics for the Arc-Routing Problem, with T. Hsu, H. Allaoui, G. Goncalves, 2014); Fahrur Rozi and Mochammad Luthfi (Airport passenger flow simulation in Quest and Arena and resource allocation optimization, with G. Gonvalves, 2011).

I created a library for Graph Coloring Benchmarking during my PhD thesis.



Teaching