Data Mining Practical Machine Learning Tools and Techniques

Constraint programming is a powerful paradigm for solving combinatorial search problems that draws on a wide range of techniques from artificial intelligence, computer science, databases, programming languages, and operations research. Constraint programming is currently applied with success to many domains, such as scheduling, planning, vehicle routing, configuration, networks, and bioinformatics.

The aim of this handbook is to capture the full breadth and depth of the constraint programming field and to be encyclopedic in its scope and coverage. While there are several excellent books on constraint programming, such books necessarily focus on the main notions and techniques and cannot cover also extensions, applications, and languages. The handbook gives a reasonably complete coverage of all these lines of work, based on constraint programming, so that a reader can have a rather precise idea of the whole field and its potential. Of course each line of work is dealt with in a survey-like style, where some details may be neglected in favor of coverage. However, the extensive bibliography of each chapter will help the interested readers to find suitable sources for the missing details. Each chapter of the handbook is intended to be a self-contained survey of a topic, and is written by one or more authors who are leading researchers in the area.

The intended audience of the handbook is researchers, graduate students, higher-year undergraduates and practitioners who wish to learn about the state-of-the-art in constraint programming. No prior knowledge about the field is necessary to be able to read the chapters and gather useful knowledge. Researchers from other fields should find in this handbook an effective way to learn about constraint programming and to possibly use some of the constraint programming concepts and techniques in their work, thus providing a means for a fruitful cross-fertilization among different research areas.

The handbook is organized in two parts. The first part covers the basic foundations of constraint programming, including the history, the notion of constraint propagation, basic search methods, global constraints, tractability and computational complexity, and important issues in modeling a problem as a constraint problem. The second part covers constraint languages and solver, several useful extensions to the basic framework (such as interval constraints, structured domains, and distributed CSPs), and successful application areas for constraint programming.

- Covers the whole field of constraint programming
- Survey-style chapters
- Five chapters on applications

Table of Contents

Foreword (Ugo Montanari) 
Part I : Foundations
Chapter 1. Introduction (Francesca Rossi, Peter van Beek, Toby Walsh) 
Chapter 2. Constraint Satisfaction: An Emerging Paradigm (Eugene C. Freuder, Alan K. Mackworth) 
Chapter 3. Constraint Propagation (Christian Bessiere) 
Chapter 4. Backtracking Search Algorithms (Peter van Beek) 
Chapter 5. Local Search Methods (Holger H. Hoos, Edward Tsang) 
Chapter 6. Global Constraints (Willem-Jan van Hoeve, Irit Katriel)
Chapter 7. Tractable Structures for CSPs (Rina Dechter)
Chapter 8. The Complexity of Constraint Languages
(David Cohen, Peter Jeavons) 
Chapter 9. Soft Constraints (Pedro Meseguer, Francesca Rossi, Thomas Schiex) 
Chapter 10. Symmetry in Constraint Programming
(Ian P. Gent, Karen E. Petrie, Jean-Francois Puget)
Chapter 11. Modelling (Barbara M. Smith) 
Part II : Extensions, Languages, and Applications

Chapter 12. Constraint Logic Programming (Kim Marriott, Peter J. Stuckey, Mark Wallace) 
Chapter 13. Constraints in Procedural and Concurrent Languages (Thom Fruehwirth, Laurent Michel, Christian Schulte)
Chapter 14. Finite Domain Constraint Programming Systems (Christian Schulte, Mats Carlsson) 
Chapter 15. Operations Research Methods in Constraint Programming (John Hooker) 
Chapter 16. Continuous and Interval Constraints(Frederic Benhamou, Laurent Granvilliers) 
Chapter 17. Constraints over Structured Domains
(Carmen Gervet) 
Chapter 18. Randomness and Structure (Carla Gomes, Toby Walsh)
Chapter 19. Temporal CSPs (Manolis Koubarakis)
Chapter 20. Distributed Constraint Programming
(Boi Faltings) 
Chapter 21. Uncertainty and Change (Kenneth N. Brown, Ian Miguel)
Chapter 22. Constraint-Based Scheduling and Planning
(Philippe Baptiste, Philippe Laborie, Claude Le Pape, Wim Nuijten)
Chapter 23. Vehicle Routing (Philip Kilby, Paul Shaw) 
Chapter 24. Configuration (Ulrich Junker)
Chapter 25. Constraint Applications in Networks
(Helmut Simonis)
Chapter 26. Bioinformatics and Constraints (Rolf Backofen, David Gilbert)

/pdf/handbook-of-constraint-programming-53srutguo2.pdf

Handbook of Constraint Programming

Fast Downward is a classical planning system based on heuristic search. It can deal with general deterministic planning problems encoded in the propositional fragment of PDDL2.2, including advanced features like ADL conditions and effects and derived predicates (axioms). Like other well-known planners such as HSP and FF, Fast Downward is a progression planner, searching the space of world states of a planning task in the forward direction. However, unlike other PDDL planning systems, Fast Downward does not use the propositional PDDL representation of a planning task directly. Instead, the input is first translated into an alternative representation called multivalued planning tasks, which makes many of the implicit constraints of a propositional planning task explicit. Exploiting this alternative representation, Fast Downward uses hierarchical decompositions of planning tasks for computing its heuristic function, called the causal graph heuristic, which is very different from traditional HSP-like heuristics based on ignoring negative interactions of operators.

In this article, we give a full account of Fast Downward's approach to solving multivalued planning tasks. We extend our earlier discussion of the causal graph heuristic to tasks involving axioms and conditional effects and present some novel techniques for search control that are used within Fast Downward's best-first search algorithm: preferred operators transfer the idea of helpful actions from local search to global best-first search, deferred evaluation of heuristic functions mitigates the negative effect of large branching factors on search performance, and multiheuristic best-first search combines several heuristic evaluation functions within a single search algorithm in an orthogonal way. We also describe efficient data structures for fast state expansion (successor generators and axiom evaluators) and present a new non-heuristic search algorithm called focused iterative-broadening search, which utilizes the information encoded in causal graphs in a novel way.

Fast Downward has proven remarkably successful: It won the "classical" (i. e., propositional, non-optimising) track of the 4th International Planning Competition at ICAPS 2004, following in the footsteps of planners such as FF and LPG. Our experiments show that it also performs very well on the benchmarks of the earlier planning competitions and provide some insights about the usefulness of the new search enhancements.

https://www.aaai.org/Papers/JAIR/Vol26/JAIR-2606.pdf

The fast downward planning system

Feel lonely? What about reading books? Book is one of the greatest friends to accompany while in your lonely time. When you have no friends and activities somewhere and sometimes, reading book can be a great choice. This is not only for spending the time, it will increase the knowledge. Of course the b=benefits to take will relate to what kind of book that you are reading. And now, we will concern you to try reading models of man as one of the reading material to finish quickly.

Models of man

When building a unified vision system or gradually adding new capabilities to a system, the usual assumption is that training data for all tasks is always available. However, as the number of tasks grows, storing and retraining on such data becomes infeasible. A new problem arises where we add new capabilities to a Convolutional Neural Network (CNN), but the training data for its existing capabilities are unavailable. We propose our Learning without Forgetting method, which uses only new task data to train the network while preserving the original capabilities. Our method performs favorably compared to commonly used feature extraction and fine-tuning adaption techniques and performs similarly to multitask learning that uses original task data we assume unavailable. A more surprising observation is that Learning without Forgetting may be able to replace fine-tuning with similar old and new task datasets for improved new task performance.

Learning without Forgetting

We present some techniques for planning in domains specified with the recent standard language PDDL2.1, supporting "durative actions" and numerical quantities. These techiques are implemented in LPG, a domain-independent planner that took part in the 3rd International Planning Competition (IPC). LPG is an incremental, any time system producing multi-criteria quality plans. The core of the system is based on a stochastic local search method and on a graph-based representation called "Temporal Action Graphs" (TA-graphs). This paper focuses on temporal planning, introducing TA-graphs and proposing some techniques to guide the search in LPG using this representation. The experimental results of the 3rd IPC, as well as further results presented in this paper, show that our techniques can be very effective. Often LPG outperforms all other fully-automated planners of the 3rd IPC in terms of speed to derive a solution, or quality of the solutions that can be produced.

/pdf/planning-through-stochastic-local-search-and-temporal-action-2fthvvtmb6.pdf

Planning through stochastic local search and temporal action graphs in LPG

Deterministic planning in the fifth international planning competition: PDDL3 and experimental evaluation of the planners

We propose an extension to the PDDL language, called PDDL3.0, that aims at a better characterization of plan quality by allowing the user to express strong and soft constraints about the structure of the desired plans, as well as strong and soft problem goals. PDDL3.0 was the reference language of the 5th International Planning competition (IPC-5). This paper contains most of the document about PDDL3.0 that was discussed by the Consulting Committee of IPC-5, and then distributed to the IPC-5 competitors. Introduction The notion of plan quality in automated planning is a practically very important issue. In many real-world planning domains, we have to address problems with a large set of solutions, or with a set of goals that cannot all be achieved. In these problems, it is important to generate plans of good or optimal quality achieving all problem goals (if possible) or some subset of them. In the previous International planning competitions, the plan generation CPU-time played a central role in the evaluation of the competing planners. In the fifth International planning competition (IPC-5), while considering the CPUtime, we would like to give greater emphasis to the importance of plan quality. The versions of PDDL used in the previous two competitions (PDDL2.1 and PDDL2.2) allow us to express some criteria for plan quality, such as the number of plan actions or parallel steps, and relatively complex plan metrics involving plan makespan and numerical quantities. These are powerful and expressive in domains that include metric fluents, but plan quality can still only be measured by plan size in the case of propositional planning. We believe that these criteria are insufficient, and we propose to extend PDDL with new constructs increasing its expressive power about the plan quality specification. The proposed extended language allows us to express strong and soft constraints on plan trajectories (i.e. constraints over possible actions in the plan and intermediate states reached by the plan), as well as strong and soft problem goals (i.e. goals that must be achieved in any valid plan, and goals that we desire to achieve, but that do not have to be necessarily achieved). Strong constraints and goals must be satisfied by any valid plan, while soft constraints and goals express desired constraints and goals, some of which may be more preferred than others. Informally, in planning with soft constraints and goals, the best quality plan should satisfy “as much as possible” the soft constraints and goals according to the specified preference relation distinguishing alternative feasible plans (satisfying all strong constraints and goals). While soft constraints have been extensively studied in the CSP literature, only very recently has the planning community started to investigate them (Brafman & Chernyavsky 2005; Briel et al. 2004; Delgrande, Schaub, & Tompits 2005; Miguel, Jarvis, & Shen 2001; Smith 2004; Son & Pontelli 2004), and we believe that they deserve more research efforts. The following are some informal examples of plan trajectory constraints and soft goals. Additional formal examples will be given in the next section. Examples in a blocksworld domain: a fragile block can never have something above it, or it can have at most one block on it; we would like that the blocks forming the same tower always have the same colour; in some state of the plan, all blocks should be on the table. Examples in a transportation domain: we would like that every airplane is used (instead of using only a few airplanes, because it is better to distribute the workload among the available resources and limit heavy usage); whenever a ship is ready at a port to load the containers it has to transport, all such containers should be ready at that port; we would like that at the end of the plan all trucks are clean and at their source location; we would like no truck to visit any destination more than once. When we have soft constraints and goals, it can be useful to give different priorities to them, and this should be taken into account in the plan quality evaluation. While there is more than one way to specify the importance of a soft constraint or goal, as a first attempt to tackle this issue, for IPC5 we have chosen a simple quantitative approach: each soft constraint and goal is associated with a numerical weight representing the cost of its violation in a plan (and hence also its relative importance with respect the other specified soft constraints and goals). Weighted soft constraints and goals are part of the plan metric expression, and the best quality plans are those optimising such an expression (more details are given in the next sections). Using this approach we can express that certain plans are more preferred than others. Some examples are (other formalised examples are given in the next sections):1 I prefer a plan where every airplane is used, rather than a plan using 100 units of fuel less, which could be expressed by weighting a failure to use all the planes by a number 100 times bigger than the weight associated with the fuel use in the plan metric; I prefer a plan where each city is visited at most once, rather than a plan with a shorter makespan, which could be expressed by using constraint violation costs penalising a failure to visit each city at most once very heavily; I prefer a plan where at the end each truck is at its start location, rather than a plan where every city is visited by at most one truck, which could be expressed by using goal costs penalising a goal failure of having every truck at its start location more heavily than a failure of having in the plan every city visited by at most one truck. We also observe that the rich additional expressive power we propose to add for goal specifications allows the expression of constraints that are actually derivable necessary properties of optimal plans. By adding them as goal conditions, we have a way to express constraints that we know will lead to the planner finding optimal plans. Similarly, one can express constraints that prevent a planner from exploring parts of the plan space that are known to lead to inefficient performance. In the next sections, we outline some extensions to PDDL2.2 that we propose for IPC-5. We call the extended language PDDL3.0. It should be noted that this is a preliminary version of the extended language, and that a more detailed description will be prepared in the future. Moreover, given that the proposed extensions are relatively new in the planning community, and that the teams participating in IPC-5 will have limited time to develop their systems, we impose some simplifying restrictions to make the language more accessible. State Trajectory Constraints Syntax and Intended Meaning State trajectory constraints assert conditions that must be met by the entire sequence of states visited during the execution of a plan. They are expressed through temporal modal operators over first order formulae involving state predicates. We recognise that there would be value in also allowing propositions asserting the occurrence of action instances in a plan, rather than simply describing properties of the states visited during execution of the plan, but we choose to restrict ourselves to state predicates in this extension of the language. The use of the extensions described here imply a new requirements flag, :constraints. The basic modal operators we propose to use in IPC-5 are: always, sometime, at-most-once, and atend (for goal state conditions). We use a special default assumption that unadorned conditions in the goal specification are automatically taken to be “at end” conditions. This assumption The benchmark domains and problems of IPC-5 contain many additional examples; some samples of them are described in (Gerevini & Long 2006). is made in order to preserve the standard meaning for existing goal specifications, despite the fact that in a standard semantics for an LTL formula an unadorned proposition would be interpreted according to the current state. We add within which can be used to express deadlines. In addition, rather than allowing arbitrary nesting of modal operators, we introduce some specific operators that offer some limited nesting. We have sometime-before, sometime-after, always-within. Other modalities could be added, but we believe that these are sufficiently powerful for an initial level of the sublanguage modelling constraints. It should be noted that, by combining these modalities with timed initial literals (defined in PDDL2.2), we can express further goal constraints. In particular, one can specify the interval of time when a goal should hold, or the lower bound on the time when it should hold. Since these are interesting and useful constraints, we introduce two modal operators as “syntactic sugar” of the basic language: hold-during and hold-after. Trajectory constraints are specified in the planning problem file in a new field, called :constraints that will usually appear after the goal. In addition, we allow constraints to be specified in the action domain file on the grounds that some constraints might be seen as safety conditions, or operating conditions, that are not physical limitations, but are nevertheless constraints that must always be respected in any valid plan for the domain (say legal constraints or operating procedures that must be respected). This also uses a section labelled (:constraints ...). The interpretation of (:constraints ...) in the conjunction of a domain and a problem file is that it is equivalent to having all the constraints added to the goals. The use of trajectory constraints (in the domain file or in the goal specification) implies the need for the :constraints flag in the :requirements list. Note that no temporal modal operator is allowed in preconditions of actions. That is, all action preconditions are with respect to a state (or time interval, in the case of overall action c

Plan Constraints and Preferences in PDDL3

The ultimate objective in planning is to construct plans for execution. However, when a plan is executed in a real environment it can encounter differences between the expected and actual context of execution. These differences can manifest as divergences between the expected and observed states of the world, or as a change in the goals to be achieved by the plan. In both cases, the old plan must be replaced with a new one. In replacing the plan an important consideration is plan stability. We compare two alternative strategies for achieving the stable repair of a plan: one is simply to replan from scratch and the other is to adapt the existing plan to the new context.

We present arguments to support the claim that plan stability is a valuable property. We then propose an implementation, based on LPG, of a plan repair strategy that adapts a plan to its new context. We demonstrate empirically that our plan repair strategy achieves more stability than replanning and can produce repaired plans more efficiently than replanning.

/pdf/plan-stability-replanning-versus-plan-repair-3aefm3ifu0.pdf

Plan stability: replanning versus plan repair

We present LPG, a fast planner using local search for solving planning graphs. LPG can use various heuristics based on a parametrized objective function. These parameters weight different types of inconsistencies in the partial plan represented by the current search state, and are dynamically evaluated during search using Lagrange multipliers. LPG's basic heuristic was inspired by Walksat, which in Kautz and Selman's Blackbox can be used to solve the SAT-encoding of a planning graph. An advantage of LPG is that its heuristics exploit the structure of the planning graph, while Black-box relies on general heuristics for SAT-problems, and requires the translation of the planning graph into propositional clauses. Another major difference is that LPG can handle action costs to produce good quality plans. This is achieved by an "anytime" process minimizing an objective function based on the number of inconsistencies in the partial plan and on its overall cost. The objective function can also take into account the number of parallel steps and the overall plan duration. Experimental results illustrate the efficiency of our approach showing, in particular, that for a set of well-known benchmark domains LPG is significantly faster than existing Graphplan-style planners.

/pdf/lpg-a-planner-based-on-local-search-for-planning-graphs-with-lbnugz9s7p.pdf

Alfonso Gerevini

Papers

Planning through stochastic local search and temporal action graphs in LPG

Deterministic planning in the fifth international planning competition: PDDL3 and experimental evaluation of the planners

Plan Constraints and Preferences in PDDL3

Plan stability: replanning versus plan repair

LPG: a planner based on local search for planning graphs with action costs