 |
Lars Kotthoff
|
I am a PhD student working on the Minion constraint solver. More specifically, I am working on a constraint solver generator. Given a constraint problem class or instance, the solver generator will generate a specialised version of Minion with regards to the algorithms and data structures used. The generation aspect is limited to the internals of the solver; a separate project, Tailor, generates an optimised Minion model of a constraint problem specified in a solver-independent modelling language.
More details can be found in my PhD proposal (pdf) and the poster (A0 pdf) I created.
My work is supervised by Ian Miguel and Ian Gent. My external SICSA supervisor is Derek Long. I am supported by a Scottish Informatics and Computer Science Alliance (SICSA) grant. I am a member of the Centre for Interdisciplinary Research in Computational Algebra (CIRCA).
Activities
I am on the local organising committee for the Constraint Programming 2010 conference.
I am taking part in the Doctoral Symposium of the Constraint Programming 2009 conference, presenting preliminary Dominion results. I have contributed a paper (PDF), a presentation (PDF), and a poster (A1 PDF).
I have written a visualiser for Minion. It shows the domains of the variables of the problem in a matrix during search. It is included with Minion starting from version 0.8. An animation for Minion solving the magic square problem for a square of size 4 is here (large gif animation).
I am tutoring CS1002 and demonstrating for CS2001 in the Martinmas semester 2009/2010. I was tutoring CS1004 in the Candlemas semester 2009 and CS1002 in the Martinmas semester 2008/2009.
Publications
-
Lars Kotthoff, Dominion -- A constraint solver generator, 15th
International Conference on Principles and Practice of Constraint
Programming Doctoral Consortium, Lisbon, Portugal, September 2009
This paper proposes a design for a system to generate constraint solvers that are specialised for specific problem models. It describes the design in detail and gives preliminary experimental results showing the feasibility and effectiveness of the approach.
Available as PDF. -
Lars Kotthoff, Constraint solvers: An empirical evaluation of design
decisions, CIRCA preprint, February 2009
This paper presents an evaluation of the design decisions made in four state-of-the-art constraint solvers; Choco, ECLiPSe, Gecode, and Minion. To assess the impact of design decisions, instances of the five problem classes n-Queens, Golomb Ruler, Magic Square, Social Golfers, and Balanced Incomplete Block Design are modelled and solved with each solver. The results of the experiments are not meant to give an indication of the performance of a solver, but rather investigate what influence the choice of algorithms and data structures has.
Available as PDF.
The analysis of the impact of the design decisions focuses on the different ways of memory management, behaviour with increasing problem size, and specialised algorithms for specific types of variables. It also briefly considers other, less significant decisions.
Other
In ancient history, I graduated from the University of Leipzig with a Diplom in Computer Science (roughly equivalent to a Master's degree). My thesis (pdf) was supervised by Gerhard Brewka and has the catchy title "Using Constraints to render Websites -- Applications of Artificial Intelligence in E-Commerce environments". After that, I spent a year in Japan on the Vulcanus programme.