In many networks, vertices have hidden attributes, or types, that are correlated with the networks topology. If the topology is known but these attributes are not, and if learning the attributes is costly, we need a method for choosing which vertex to query in order to learn as much as possible about the attributes of the other vertices. We assume the network is generated by a stochastic block model, but we make no assumptions about its assortativity or disassortativity. We choose which vertex to query using two methods: 1) maximizing the mutual information between its attributes and those of the others (a well-known approach in active learning) and 2) maximizing the average agreement between two independent samples of the conditional Gibbs distribution. Experimental results show that both these methods do much better than simple heuristics. They also consistently identify certain vertices as important by querying them early on.

Luhmann (1984) defined society as a communication system which is structurally coupled to, but not an aggregate of, human action systems. The communication system is then considered as self-organizing ("autopoietic"), as are human actors. Communication systems can be studied by using Shannon's (1948) mathematical theory of communication. The update of a network by action at one of the local nodes is then a well-known problem in artificial intelligence (Pearl 1988). By combining these various theories, a general algorithm for probabilistic structure/action contingency can be derived. The consequences of this contingency for each system, its consequences for their further histories, and the stabilization on each side by counterbalancing mechanisms are discussed, in both mathematical and theoretical terms. An empirical example is elaborated.

Recent research on multiple kernel learning has lead to a number of approaches for combining kernels in regularized risk minimization. The proposed approaches include different formulations of objectives and varying regularization strategies. In this paper we present a unifying general optimization criterion for multiple kernel learning and show how existing formulations are subsumed as special cases. We also derive the criterion's dual representation, which is suitable for general smooth optimization algorithms. Finally, we evaluate multiple kernel learning in this framework analytically using a Rademacher complexity bound on the generalization error and empirically in a set of experiments.

This paper constructively proves the existence of an effective procedure generating a computable (total) function that is not contained in any given effectively enumerable set of such functions. The proof implies the existence of machines that process informal concepts such as computable (total) functions beyond the limits of any given Turing machine or formal system, that is, these machines can, in a certain sense, "compute" function values beyond these limits. We call these machines creative. We argue that any "intelligent" machine should be capable of processing informal concepts such as computable (total) functions, that is, it should be creative. Finally, we introduce hypotheses on creative machines which were developed on the basis of theoretical investigations and experiments with computer programs. The hypotheses say that machine intelligence is the execution of a self-developing procedure starting from any universal programming language and any input.

One of the objectives of designing feature selection learning algorithms is to obtain classifiers that depend on a small number of attributes and have verifiable future performance guarantees. There are few, if any, approaches that successfully address the two goals simultaneously. Performance guarantees become crucial for tasks such as microarray data analysis due to very small sample sizes resulting in limited empirical evaluation. To the best of our knowledge, such algorithms that give theoretical bounds on the future performance have not been proposed so far in the context of the classification of gene expression data. In this work, we investigate the premise of learning a conjunction (or disjunction) of decision stumps in Occam's Razor, Sample Compression, and PAC-Bayes learning settings for identifying a small subset of attributes that can be used to perform reliable classification tasks. We apply the proposed approaches for gene identification from DNA microarray data and compare our results to those of well known successful approaches proposed for the task. We show that our algorithm not only finds hypotheses with much smaller number of genes while giving competitive classification accuracy but also have tight risk guarantees on future performance unlike other approaches. The proposed approaches are general and extensible in terms of both designing novel algorithms and application to other domains.

We consider the problem of reinforcement learning using function approximation, where the approximating basis can change dynamically while interacting with the environment. A motivation for such an approach is maximizing the value function fitness to the problem faced. Three errors are considered: approximation square error, Bellman residual, and projected Bellman residual. Algorithms under the actor-critic framework are presented, and shown to converge. The advantage of such an adaptive basis is demonstrated in simulations.

We present an extension of the planning framework based on action graphs and local search to deal with PDDL2.1 temporal problems requiring concurrency, while previously the approach could only solve problems admitting a sequential solution. The paper introduces a revised plan representation supporting concurrency and some new search techniques using it, which are implemented in a new version of the LPG planner. An experimental analysis indicates that the proposed approach is suitable to temporal planning with requiring concurrency and is competitive with state-of-the-art planners.

Compared to optimal planners, satisficing planners can solve much harder problems but may produce overly costly and long plans. Plan quality for satisficing planners has become increasingly important. The most recent planning competition IPC-2008 used the cost of the best known plan divided by the cost of the generated plan as an evaluation metric. This paper proposes and evaluates two simple but effective methods for plan improvement: Action Elimination improves an existing plan by repeatedly removing sets of irrelevant actions. Plan Neighborhood Graph Search finds a new, shorter plan by creating a plan neighborhood graph PNG(π) of a given plan π, and then extracts a shortest path from PNG(π). Both methods are implemented in the Aras postprocessor and are empirically shown to improve the result of several planners, including the top four planners from IPC-2008, under competition conditions.

Disaster Response Robotics is a challenging domain, where distributed. New challenges hence come up in terms the need for intelligent robotic agents (as opposed to just of autonomy, cooperation and collective behaviors. In the robots) is motivated both by technical considerations and in first part of the talk, I briefly overview the state of the art a practical application perspective. In emergency scenarios in the field of disaster response robotics, in order to support time is critical. Hence, there is a great demand for tools the above sketched analysis. In the second part of that improve the effectiveness of operations. Although there the talk, I present some of the research we developed at are specific actions that can be accomplished by a robot, Sapienza Univ. of Rome, also in collaboration with the Italian such as for example bomb disposal, a key goal of disaster Firemen Department. Specifically, I describe some results response robots is to acquire knowledge about the scenario.

In this talk, I will introduce computer-aided algorithm design and discuss its main ingredients: design patterns, which provide ways of structuring potentially large spaces of candidate algorithms, and meta-algorithmic optimisation procedures, which are used for finding good designs within these spaces. After explaining how this algorithm design approach differs from and complements related approaches in program synthesis, genetic programming and so-called hyperheuristics, I will illustrate its success using examples from our own work in SAT-based software verification (Hutter et al. 2007), timetabling (Chiarandini, Fawcett, and Hoos 2008) and mixed integer programming (Hutter, Hoos, and Leyton-Brown 2010). Furthermore, I will argue why this approach can be expected to be particularly useful and effective for building better solvers for rich and diverse classes of combinatorial problems, such as planning and scheduling. Finally, I will outline out how programming by optimisation — a design paradigm that emphasises the automated construction of performance-optimised algorithm by means of searching large spaces of alternative designs — has the potential to transform the design of high-performance algorithm from a craft that is based primarily on experience and intuition into a principled and highly effective engineering effort.