I got my Ph.D. in Industrial and Information Engineering on April 4, 2014, at the University of Udine. My main research interests are propagation-based hybrid meta-heuristics for combinatorial optimization, and sustainable mobility problems. I'm also exploring machine learning techniques (reinforcement learning, artificial neural networks) as a way to mix together different search methods.
During my Ph.D., I have spent ~8 months in Adelaide, Australia working on Multi-Objective Optimization algorithms and runtime analysis of Genetic Programming.
In December, 2013 I submitted my Ph.D. thesis for approval. Here is the pre-print version (March, 2014).
- Urli, T. (2014). Hybrid meta-heuristics for combinatorial optimization. PDF.
Here is an updated list of my publications, organized per research field and sorted by year (desc).
Meta-heuristics, hybrid-heuristics and hyper-heuristics
- Battistutta, M., Schaerf, A., Urli, T. (2014). Feature-based tuning of single-stage simulated annealing for examination timetabling. In Proceedings of PATAT'14: 10th International Conference on the Practice and Theory of Automated Timetabling | PDF.
- Di Gaspero, L. & Urli, T. (2014). A CP/LNS approach for Multi-day Homecare Scheduling Problems. In Proceedings of HM'14: 9th International Workshop on Hybrid Metaheuristics | PDF | Slides | Talk recording.
- Di Gaspero, L. & Rendl, A. & Urli, T. (2015). Balancing Bike Sharing Systems with Constraint Programming. Constraints | PDF.
- Urli, T. (2013). Hybrid CP+LNS for the Curriculum-Based Course Timetabling Problem. In CP'13 Doctoral Program | PDF | Slides.
- Di Gaspero, L. & Rendl, A. & Urli, T. (2013). Constraint-based approaches for Balancing Bike Sharing Systems . In Proceedings of CP'13: the 19th International Conference on Principles and Practice of Constraint Programming | PDF | Code.
- Bellio R. & Ceschia S. & Di Gaspero L. & Schaerf A. & Urli T. (2013). A simulated annealing approach to the curriculum-based course timetabling problem . In Proceedings of MISTA'13: the 6th Multidisciplinary International Scheduling Conference: Theory and Applications | PDF | Slides.
- Di Gaspero, L. & Rendl, A. & Urli, T. (2013). A Hybrid ACO+CP for Balancing Bicycle Sharing Systems . In Proceedings of HM'13: 8th International Workshop on Hybrid Metaheuristics | PDF.
- Di Gaspero, L. & Urli, T. (2012). Evaluation of a family of Reinforcement Learning cross-domain heuristics for optimization. In Proceedings of LION 6: Learning and Intelligent Optimization | PDF | Normalized function value plots.
- Di Gaspero, L. & Urli, T. (2011). A Reinforcement Learning approach for the Cross-Domain Heuristic Search Challenge. In Proceedings of MIC 2011: The IX Metaheuristics International Conference | PDF | Slides.
In 2011 participated in the Cross-domain Heuristic Search Challenge 2011 (CHeSC 2011) with a hyper-heuristics based on reinforcement learning. My hyper-heuristic is ranked 16th. A research paper on this subject has been recently published at the MIC 2011: IX Metaheuristics International Conference; I gave a talk on the topic.
Tools for Experimental Analysis
During my visiting period at the Evolutionary Computation Group at the University of Adelaide I have been working with Prof. Frank Neumann, Mr. Markus Wagner and Mr. Anh Quang Nguyen on the analysis of Genetic Programming systems on simple optimization problems (order, majority, sorting). Following are some of the results of this research.
- Wagner M., Neumann F., T. Urli. (2015). On the Performance of Different Genetic Programming Approaches for the SORTING Problem. In Evolutionary Computation Journal (MIT Press). (Accepted).
- Urli T., Wagner M. and Neumann F. (2012). Experimental Supplements to the Computational Complexity Analysis of Genetic Programming for Problems Modelling Isolated Program Semantics. In Proceedings of PPSN 2012 - 12th International Conference on Parallel Problem Solving From Nature | PDF.
- Nguyen A. Q., Urli T. and Wagner M. (2013). Single- and Multi-Objective Genetic Programming: New Bounds for Weighted Order and Majority. In Post-proceedings of FOGA 2013 - Foundation of Genetic Algorithms 2013 | PDF
- GPFramework (0.1) a Java framework for rapid prototyping of mutation-only Genetic Programming systems (mostly, but not exclusively, with the goal to assess their computational complexity). Hosted at:
Virtual camera control
Even if, strictly speaking, it has never been published anywhere, my M.Sc. thesis (English) is about virtual camera control. I also authored a paper on the topic
- Ranon, R., Christie, M. & Urli, T. (2010). Accurately Measuring the Satisfaction of Visual Properties in Virtual Camera Control. In R. Taylor, P. Boulanger, A. Krüger & P. Olivier (Eds.), Proceedings of Smart Graphics 2010 (Vol. 6133, pp. 91-102). Berlin, Heidelberg: Springer Berlin Heidelberg. | PDF.
- Ranon, R. & Urli, T. (2014). Improving the Efficiency of Viewpoint Computation. In IEEE Transactions on Visualization and Computer Graphics (Issue 99). | PDF.
I am also involved in various activities related to the academic world
Every now and then I give some extra lectures (e.g. laboratory hours) for courses at the Faculty of Engineering. Here are some slides about practical network and system attacks I gave in 2011 for the Computer security course by Prof. Pier Luca Montessoro.
- Lecture about buffer overflow and SQL injection | Slides | Material,
- lecture about mitm attacks over LAN networks | Slides | Material (for patched Ettercap, see license in package),
- lecture about WEP networks cracking | Slides
Note: all the slides are in Italian.
In case you need to contact me you can either:
- dial +61 6267 6362,
- drop me a line at ,
- come find me at
Tower A, Level 3
7 London Circuit Canberra, ACT 2601