[var] => content
[type] => preview
[var] => cut_url
National ICT Australia, University of Melbourne
Barry Street, Level
• Researcher, Constraint Programming Platform G, National ICT Australia NICTA, Victoria Re
search Laboratory, Melbourne, Australia, August present.
• Honorary Research Fellow, University of Melbourne, Melbourne, Australia, August present.
• PostDoctoral Researcher, Constraint Reasoning and Machine Learning Group, Microsoft Research,
Cambridge, England, September July .
• Research Internship, Constraint Reasoning Group, Microsoft Research, Cambridge, England,
July October .
• Research Assistant, Artificial Intelligence and Knowledge Representation Group, University of Toronto,
Canada, September August .
• Ph.D. Computer Science, University of Toronto, Artificial Intelligence Group. Advisor: Professor
Fahiem Bacchus. Project: Quantified Boolean Formulas, September August
• M.Sc. Computer Science Minor: Neurobiology, University of Technology RWTH, Aachen, Ger
many, Knowledge Representation Group. Advisor: Professor Gerhard Lakemeyer. Project: Evaluation
Based Reasoning in FirstOrder Knowledge Bases, October February
Refereed Conference Publications
• [AAAI] David Stern, Horst S., Ralf Herbrich, Thore Graepel: Collaborative Portfolio
Manager, AAAI , The TwentyFourth AAAI Conference on Artificial Intelligence, submitted
• [IJCAI] Lucas Bordeaux, Youssef Hamadi, Horst S.: Experiments with Massively
Parallel Constraint Solving, IJCAI , International Joint Conference on Artificial Intelligence,
• [AAAI] Horst S., Roland Memisevic: Learning to Solve QBF, AAAI , The Twenty
Second AAAI Conference on Artificial Intelligence,
• [SAT] Horst S., Fahiem Bacchus: Dynamically Partitioning for Solving QBF, SAT
, Tenth International Conference on Theory and Applications of Satisfiability Testing,
• [SAC] Lucas Bordeaux, Horst S.: On the Stochastic Constraint Satisfaction Frame
work, SAC, The nd Annual ACM Symposium on Applied Computing,
• [CP] Horst S., Jessica Davies, Fahiem Bacchus: Preprocessing QBF, CP , Prin
ciples and Practice of Constraint Programming, Springer New York,
• [SAT] Horst S., Fahiem Bacchus: Binary Clause Reasoning in QBF, SAT , Ninth
International Conference on Theory and Applications of Satisfiability Testing, Springer New York,
• [CP] Horst S., Fahiem Bacchus: Using SAT in QBF, CP, Principles and Practice
of Constraint Programming, Springer New York,
• [GI] Alexander Hornung, Horst S.: DVisualization of Music, German Society for
Computer Science GI, Computer Science Days, Ulm,
• [CGI] Walter Oberschelp, Alexander Hornung, Horst S.: Visualization of Eclipses and
Planetary Conjunction Events: The Interplay between Model Coherence, Scaling and Anima
tion, CGI , Computer Graphics International, Geneva,
Refereed Workshop Publications
• [LION] Yuehua Xu, David Stern, Horst S.: Learning Adaptation to Solve Constraint
Problems, short paper, LION , Learning and Intelligent OptimizatioN,
• [NESCAI] Jessica Davies, Eric Hsu, Horst S.: Solution Backtracking for SAT,
NESCAI , North East Student Colloquium on Artificial Intelligence,
• [NESCAI] Horst S., Fahiem Bacchus: Updated and compacted version of Using SAT
in QBF, NESCAI , North East Student Colloquium on Artificial Intelligence,
• [NESCAI] Horst S., Jessica Davies, Fahiem Bacchus: Preliminary version of Prepro
cessing QBF, NESCAI , North East Student Colloquium on Artificial Intelligence,
• [JV C] Walter Oberschelp, Alexander Hornung, Horst S.: Visualization of eclipses and
planetary conjunction events. The interplay between model coherence, scaling and animation.
The Visual Computer, Vol. : ,
• Horst S., Solving Quantified Boolean Formulas, PhD Thesis, University of Toronto,
• Horst S., The Efficiency and Implementation of an EvaluationBased Reasoning Pro
cedure with Disjunctive Information in FirstOrder Knowledge Bases, Master Thesis, RWTH
• Collaborative Expert Portfolio Management at Microsoft Research
This work enables to manage a portfolio of experts such as automated problem solvers, algorithms,
automated services or human experts in order to select the most appropriate expert for a given task.
• Contributing to the G Constraint Reasoning Platform
• Contributed to the Microsoft Solver Foundation
• Developed an extensible and lightweight searchbased QBF solver called MiniQBF written in C++
• The following software to solve Quantified Boolean Formulas QBF is available:
– QBF Preprocessor PreQuel
– QBF Solver SQBF
– QBF Solver clsQ
The software mostly developed in Microsoft Visual Studio can be found at:
Other Work Experience
• Founder and Executive Board, ITDEVELOP, Aachen, Germany, May January . The
main activities were related to the development of a contemporary communication platform that en
abled interactive locationbased event management. Mobility is becoming evermore important and
present in our society both in business and in private life. People use mobile phones, PDAs and
laptop computers, drive to work and to events, fly to business meetings and on holidays. Although
mobility is increasing and peoples geographical horizons are broadening, information sources did not
• Editor, March . German translation of the book ”The Essence of Artificial Intelligence” by Alison
• Research Assistant, Virtual Reality Lab, RWTH Aachen, Germany, February August .
The work mainly concentrated on developing a virtual reality software that supported multiple pro
jectors and operated on a distributed system Linux Cluster. This work was presented at several
conferences and computer trade fairs e.g., CeBIT Hannover.
• Software Developer and System Administrator, HEWATT Corp., Stolberg, Germany, .
The software projects were mainly focused on embedded systems and the simulation of prototypes
e.g., blood pressure meter. The duties as a system administrator included the installation and sup
port of Windows/Linux machines and networking SAMBA.
Teaching and Mentoring
• CoSupervision of summer student, NICTA,
Topic: Partial Booleanization of constraints defined in more expressive modelling languages e.g.,
• Leading Internship in the Constraint Reasoning Group at Microsoft Research,
Topic: Using Reinforcement Learning to dynamically adapt search when solving Constraint Problems.
• Teaching Assistant for The How and Why of Computing CSC , University of Toronto, Spring
Topics: HTML, JAVA, Algorithms and Data Structures, Database Systems, Theory of Computation
Duties included marking assignments and exams, conducting weekly one hour classroom tutorials/lab
sessions, maintaining newsgroup and also setting up assignments, tutorials, and lab questions.
• Mentoring Undergraduate and Master students at the University of Toronto,
• Research Group Organizer, Knowledge Representation Group, University of Toronto,
• Teaching Assistant for Introduction to Artificial Intelligence CSC , University of Toronto, Win
ter , Spring , Winter , Spring , Winter , Spring , Winter .
Topics: What is AI, Search e.g., Backtracking/Game Tree Search, Knowledge Representation, Rea
soning Under Uncertainty, Planning
Duties included marking assignments and exams, conducting weekly one hour classroom tutorials,
office hours, and also setting up assignments/exams.
Academic Awards, Scholarships and Funding
• Winner of the International QBF Competition
st, nd, and rd place on the industrial benchmarks see for more details
• PhD. Scholarship, ≈ CAD, , University of Toronto,
• Interestfree Loan for StartUp, ≈ EUR, , Sparkasse Aachen, Germany,
• Computer Languages: C, C, C++, F, Python, Java, HTML, PHP, SQL, Prolog, Mercury
• Familiar with HighPerformance Computing Clusters Microsoft Windows HPC and Linux clusters
• Fluent in English, German, and Dutch; at an intermediate level in Spanish; basic knowledge in French.