Home » Archivo » archive 2010/2011

Archive 20102011

Jump second level menu

Corsi Laurea

Corsi Laurea Magistrale

Seminari

  • Distributed Detection of Clone Attacks in Wireless Sensor Networks - 2011-10-07 - Mauro Conti (Universita' di Padova)

    Abstract: Wireless Sensor Networks (WSNs) are often deployed in hostile environments where an adversary can physically capture some of the nodes, first can reprogram, and then, can replicate them in a large number of clones, easily taking control over the network. A few distributed solutions to address this fundamental problem have been recently proposed. However, these solutions are not satisfactory. First, they are energy and memory demanding: A serious drawback for any protocol to be used in the WSN-resource-constrained environment. Further, they are vulnerable to the specific adversary models that we introduced. The contributions of our work are threefold. First, we analyze the desirable properties of a distributed mechanism for the detection of node replication attacks. Second, we show that the known solutions for this problem do not completely meet our requirements. Third, we propose a new self-healing, Randomized, Efficient, and Distributed (RED) protocol for the detection of node replication attacks, and we show that it satisfies the introduced requirements. Finally, extensive simulations show that our protocol is highly efficient in communication, memory, and computation; is much more effective than competing solutions in the literature; and is resistant to the new kind of attacks introduced in our work, while other solutions are not.

  • Random projections and dynamics for robot motion learning: From extreme learning machines to reservoir computing. - 2011-09-12 - Prof. Dr. Jochen Steil (CoR-Lab, Bielefeld University, Germany )

    Abstract: Random feature generation for data processing is a recently popular approach to create fast learning schemes in various neural network architectures. Starting from a simple extreme learning machine (ELM), it will be shown how to use regularization and feedback to improve the feature generation for regression and classification tasks with static data. Adding feedback then leads to recurrent reservoir networks which are capable of learning also temporal data and a new approach for hidden state associative learning. I will demonstrate these methods for new motion learning applications (learning inverse kinematics, trajectory generation, sensorimotor transforms) including the representation of multiple redundant motor patterns in one network. Application examples will include the child-like iCub, the Honda humanoid robot, and a system for interactive programming of the compliant DLR-Kuka robot arm.

  • Recent Advances in Solving Combinatorial Optimization Tasks over Graphical Models - 2011-09-09 - Rina Dechter (UC Irvine, USA)

    Abstract: In this talk I will present state of the art algorithms for solving combinatorial optimization tasks defined over graphical models (Bayesian networks, Markov networks, and constraint networks) and demonstrate their performance on a variety of benchmarks. Specifically I will present branch and bound and best-first search algorithms which explore the AND/OR search space over graphical models and will demonstrate the gain obtained by exploiting problem’s decomposition (using AND nodes), equivalence (by caching) and irrelevance (via the power of new lower bound heuristics such as mini-buckets). The impact of additional principles such as exploiting determinism via constraint propagation, the use of good initial upper bounds generated via stochastic local search and the variable orderings ideas may be discussed, as time permits. This is joint work with Radu Marinescu.

  • Influence, Bribery, and Manipulation in Voting Systems - 2011-06-16 - Nicholas Mattei (University of Kentucky, Lexington, Kentucky)

    Abstract: Computational Social Choice (ComSoc) is an emergent and vibrant area of research in Computer Science. ComSoc, in broad terms, is concerned with the design and analysis of systems and processes for collective decision making. Voting and election procedures are common ways that groups of agents can arrive at a collective decision. Unfortunately, foundational results in the field of Social Choice prove that it is impossible to devise a voting procedure for three or more candidates that is immune to manipulation (some agent will, in some cases, have an incentive to misrepresent his true preferences). Our research focuses on the manipulation as well as bribery problems in voting procedures. Most research related to bribery and manipulation assumes an agent has access to perfect information about the preferences of all agents within the system. Our research focuses on the case where an agent only has access to probabilistic information about other agents’ preferences. This talk will provide a brief introduction to ComSoc, a review of some major results related to election systems, and an overview of our work on election systems where voters’ preferences are modeled as probabilities.

  • Experiments in Parallel Constraint-Based Local Search - 2011-05-27 - Philippe Codognet (University Pierre et Marie Curie (Paris 6), France )

    Abstract: We present a parallel implementation of a constraint-based local search algorithm and investigate its performance on hardware with several hundreds of processors. We choose as constraint solving algorithm for these experiments the adaptive search method, an efficient sequential local search method for solving CSPs. The implemented algorithm is a parallel version of adaptive search in a multiple independent-walk manner, that is, each process is an independent search engine and there is no communication between the simultaneous computations. Performance evaluation on a variety of classical CSPs benchmarks shows that speedups are very good for a few tens of processors, and good up to a few hundreds of processors. We also investigated more complex parallelization schemes with communication between processors and will report about early experimental results.

  • What is a Cluster? Perspectives from Game Theory - 2011-03-03 - Marcello Pelillo (Universita' Ca' Foscari, Venezia )

    Abstract: Contrary to the vast majority of approaches to data clustering, which view the problem as one of partitioning a set of observations into coherent classes, thereby obtaining the clusters as a by-product of the partitioning process, we propose to reverse the terms of the problem and attempt instead to derive a rigorous formulation of the very notion of a cluster. In our endeavor to provide an answer to this question, we found that game theory offers an elegant and general perspective that serves well our purposes. Accordingly, we formulate the clustering problem as a non-cooperative 'clustering game'. Within this context, the notion of a cluster turns out to be equivalent to the concept of evolutionary stable strategy (ESS), a classical equilibrium concept from (evolutionary) game theory, of which we offer a fresh combinatorial characterization. Computationally, ESS-clusters can be found using, e.g., replicator dynamics, a well-known formalization of natural selection processes. The proposed framework has found applications in a variety of application fields, including bioinformatics, computer vision, image processing, security and video surveillance, etc. In the talk, I will be presenting some experimental results on image segmentation and related problems.

  • Molti modi di dire causa. Ovvero, i sistemi concorrenti han bisogno della causalita’? - 2011-02-14 - Dr F. Russo (Philosophy - SECL, University of Kent (UK) )

    Abstract: I questo seminario si presenteranno i principali approcci alla causalita', dal regolarismo humeano ai contemporanei dibattiti sul 'difference-making' e sui meccanismi. Si distingueranno anche diversi approcci alla causalita', ovvero come studio del linguaggio (causale) ordinario o scientifico, oppure della pratica scientifica. Si affrontera' infine la questione del perche' interessarsi alla ricerca delle cause nella scienza, dalla comprensione di un fenomeno alla predizione delle politiche sociali od economiche.

  • Vehicular Congestion Detection and Short-Term Forecasting: A New Model with Results - 2010-12-02 - Gustavo Marfia (Università degli Studi di Bologna)

    Abstract: While vehicular congestion is very often defined in terms of aggregate parameters such as traffic volumes and lane occupancies, based on recent findings the interpretation that receives most credit is that of a state of a road where traversing vehicles experience a delay exceeding the maximum value typically incurred under light or free-flow traffic conditions. We here propose a new definition, according to which a road is in a congested state (be it high or low) only when the likelihood of finding it in the same congested state is high in the near future. Based on this new definition, we devised an algorithm which, exploiting probe vehicles, for any given road: (a) identifies if it is congested or not, and; (b) provides the estimation that a current congested state will last for at least a given time interval. Unlike any other existing approach, an important advantage of ours is that it can be generally applied to any type of road, as it neither needs any a-priori knowledge nor require to estimate any road parameter (e.g., number of lanes, traffic light cycle, etc.). Further, it allows to perform short term traffic congestion forecasting for any given road. We present several field trials gathered on different urban roads whose empirical results confirm the validity of our approach.

  • Informatica Pediatrica per interventi riabilitativi precoci - 2010-11-22 - Roberto Mancin (Dipartimento di Pediatria - Università di Padova)

    Abstract: L'Informatica Pediatrica è la branca delle scienze informatiche che si occupa di massimizzare il livello di maturazione e di sviluppo di OGNI persona in età evolutiva tramite opportune tecnologie digitali. Queste, se utilizzate precocemente, prevengono comportamenti disadattivi e favoriscono i processi riabilitativi.