We live in a period where Information and Communication Technologies (ICT) has revolutionized the way we communicate, learn, work, and entertain ourselves. But we also live in challenging times, from climate change and natural disasters of increased intensity to rapid urbanization, pollution, economic stagnation, and a shrinking middle class in Western countries. In this lecture, we argue that ICT now has the opportunity to radically change the way we take decisions as a society, exploiting the wealth of data available to understand physical, biological, business, and human behaviors with unprecedented accuracy and speed. We illustrate this vision with challenging problems in disaster management, energy, medicine, and transportation.
Van Hentenryck leads the Optimization Research Group at NICTA, whose research focuses on optimization, algorithmic decision theory, logistics and supply chains, energy systems, and disaster management. He is also a professor at the University of Melbourne.
Van Hentenryck is the recipient of two Honorary degrees, the 2002 Informs ICS Award for research excellence at the intersection of operations research and computer, the 2006 ACP Award for research excellence in constraint programming, the 2010-2011 Philip J. Bray Award for Excellence in Undergraduate Teaching at Brown University. He is a 2013 IFORS distinguished lecturer and a fellow of the Association for the Advancement of Artifical Intelligence. Van Hentenryck is the author of five MIT Press books and has developed a number of innovative optimization systems that are widely used in academia and industry. His research on disaster management has been deployed to help federal agencies in the United States mitigate the effects of hurricanes on coastal areas.
Optimization for policy making in the energy sector: the cornerstone for an integrated approach
In this talk I will present a recent research activity on the use of constraint programming and optimization techniques for supporting policy making in the energy sector. Policy making is a very complex task taking into account several aspects related to sustainability, namely impact on the environments, health of productive sectors, economic implications and social acceptance. I will show that optimization methods could be extremely useful for analysing alternative policy scenarios, but should be complemented with several other techniques such as machine learning, agent-based simulation, opinion mining and visualization to come up with an integrated system able to support decision making in the overall policy design life cycle. I will discuss how these techniques could be merged with optimization and I will identity some open research directions.
Michela Milano is associate professor at ISI, the Department of Computer Science and Engineering of the University of Bologna. She received her Ph.D. in Computer Science in 1998. Her research interests cover the area of hybrid optimization, optimization for embedded system design and computational sustainability. She is author of more than 120 papers on peer reviewed international conferences and journals, editor of five books on hybrid optimization and guest editor of six special issues. She is one of the founders of the CP and Operations Research community, program chair of CPAIOR 2005 and CPAIOR 2010. She has been program chair of CP 2012. She is member of the program committee of the main conferences in the field, member of the Editorial Board of the Constraint Journal, Area Editor of Constraint Programming Letters and Area Editor of INFORMS Journal on Computing.
Michela Milano is coordinator of the EU FP7 project e-POLICY - Engineering the POlicy making LIfe CYcle (2011-2014), aimed at providing decision support systems for policy makers. She is partner of the COLOMBO - Cooperative Self-Organizing system for low carbon mobility at low penetration rates (2012-2015). She has been Italian coordinator for an exchange programme Italy Quebec: Algorithms and systems for the operational planning in industry and services (2007-2009) and recipient of the Google Focused Grant Program on Mathematical Optimization and Combinatorial Optimization in Europe on Model Learning in Combinatorial Optimization (2012-2013).
Answer Set Programming: Boolean Constraint Solving for Knowledge Representation and Reasoning
Answer Set Programming (ASP) is a declarative problem solving approach, combining a rich yet simple modeling language with high-performance Boolean constraint solving capacities. ASP is particularly suited for modeling problems in the area of Knowledge Representation and Reasoning involving incomplete, inconsistent, and changing information. As such, it offers, in addition to satisfiability testing, various reasoning modes, including different forms of model enumeration, intersection or unioning, as well as multi-criteria and -objective optimization. From a formal perspective, ASP allows for solving all search problems in NP (and NP^NP) in a uniform way. Hence, ASP is well-suited for solving hard combinatorial search problems, like system design and timetabling. Prestigious applications of ASP include composition of Renaissance music, decision support systems for NASA shuttle controllers, reasoning tools in systems biology and robotics, industrial team-building, and many more. The versatility of ASP is nicely reflected by the ASP solver clasp, winning first places at various solver competitions, such as ASP, MISC, PB, and SAT competitions. The solver clasp is at the heart of the open source platform Potassco hosted at potassco.sourceforge.net. Potassco stands for the "Potsdam Answer Set Solving Collection" and has seen more than 30000 downloads world-wide since its inception at the end of 2008.
The talk will start with an introduction to ASP, its modeling language and solving methodology, and portray some distinguished ASP systems.
Torsten Schaub is University Professor at the University of Potsdam, Germany. He received his dissertation in 1992 from the Technical University of Darmstadt, Germany and his habilitation in 1995 from the University of Rennes I, France. He is also an Adjunct Professor at the School of Computing Science at Simon Fraser University, Canada, and at the Institute for Integrated and Intelligent Systems at Griffiths University, Australia. Since 2012, Torsten Schaub is a fellow of ECCAI. His research focuses on the automatisation of reasoning from incomplete, contradictory, and evolutive information. His research interests range from theoretic foundations to practical implementations, with a focus on Answer set programming (ASP). The latter activity has led to the open source project potassco.sourceforge.net, bundling tools for ASP developed at Potsdam. Potassco comprises more than a dozen ASP-related systems, among them the award winning solver clasp.
"Those who cannot remember the past are condemned to repeat it"
Constraint programming is a highly successful technology for tackling complex combinatorial optimization problems. Any form of combinatorial optimization involves some form of search, and CP is very well adapted to make use of programmed search and strong inference to solve some problems that are out of reach of competing technologies. But much of the search that happens during a CP execution is effectively repeated. This arises from the combinatorial nature of the problems we are tackling. Learning about past unsuccessful searches and remembering this in an effective way can exponentially reduce the size of the search space. In this talk I will explain lazy clause generation, which is a hybrid constraint solving technique that steals all the best learning ideas from Boolean satisfiability solvers, but retains all the advantages of constraint programming. Lazy clause generation provides the state of the art solutions to a wide range of problems, and consistently outperforms other solving approaches in the MiniZinc challenge
Peter J. Stuckey is a Professor in the Department of Computing and Information Systems in the University of Melbourne, and project leader in the National ICT Australia Victoria laboratory. Peter Stuckey is a pioneer in constraint programming, the science of modelling and solving complex combinatorial problems. His research interests include: constraint programming; programming languages, in particular declarative programming languages; constraint solving algorithms; bioinformatics; and constraint-based graphics. He enjoys problem solving in any area, having publications in e.g. databases, timetabling, and system security.
Peter Stuckey received a B.Sc and Ph.D both in Computer Science from Monash University in 1985 and 1988 respectively. In 2009 he was recognized as an ACM Distinguished Scientist. In 2010 he was awarded the Google Australia Eureka Prize for Innovation in Computer Science for his work on lazy clause generation. He was awarded the 2010 University of Melbourne Woodward Medal for most outstanding publication in Science and Technology across the university.
The ObjectiveCP Optimization System
ObjectiveCP is an optimization system that views an optimization program as the combination of a model, a search, and a solver. Models in ObjectiveCP follow the modeling style of constraint programming and are concretized into specific solvers. Search procedures are specified in terms of highlevel nondeterministic constructs, search combinators, and node selection strategies. ObjectiveCP supports fully transparent parallelization of multistart and branch & bound algorithms. The implementation of ObjectiveCP is based on a sequence of model transformations, followed by a concretization step. Moreover, ObjectiveCP features a constraint programming solver following a microkernel architecture for ease of maintenance and extensibility. Experimental results show the practicability of the approach.
Pascal leads the Optimization Research Group at NICTA, whose research focuses on optimization, algorithmic decision theory, logistics and supply chains, energy systems, and disaster management. He is also a professor at the University of Melbourne.
Laurent is an Associate Professor in the Computer Science & Engineering Department at the University of Connecticut. His research interests lie primarily with systems for Constraint Programming, Constraint-Based Local Search and Solvers in general. He co-authored with Pascal several systems including Numerica, Localizer, OPL and COMET.