### Lista dei rapporti tecnici dell'anno 2011

With the support of the legally-grounded methodology of situation
testing, we tackle the problems of discrimination discovery and
prevention from a dataset of historical decisions by adopting a
variant of k-NN classification. A tuple is labeled as
discriminated if we can observe a significant difference of
treatment among its neighbors belonging to a protected-by-law
group and its neighbors not belonging to it. Discrimination
discovery boils down to extracting a classification model from the
labeled tuples. Discrimination prevention is tackled by changing
the decision value for tuples labeled as discriminated before
training a classifier. The approach of this paper overcomes legal
weaknesses and technical limitations of existing proposals.

Aldinucci, Marco and Ruggieri, Salvatore and Torquati, Massimo Porting Decision Tree Building and Pruning Algorithms to Multicore using FastFlow December 4, 2014 UnipiEprints view Download (840Kb) abstract The whole computer hardware industry embraced multicores. For these machines, the extreme optimisation of sequential algorithms is no longer sufficient to squeeze the real machine power, which can be only exploited via thread-level parallelism. Decision tree algorithms exhibit natural concurrency that makes them suitable to be parallelised. This paper presents an approach for easy-yet-efficient porting of an implementation of the C4.5 algorithm on multicores. The approach is based on the FastFlow parallel programming environment. The strength of our porting consists in minimal changes to the original sequential code. In addition to the tree building algorithm, we consider also the so far unaddressed problem of parallelising the error-based pruning with grafting algorithm of C4.5. We devise lower bounds for the forms of parallelisations adopted, and achieve performances close to such bounds. Bigi, Giancarlo and Passacantando, Mauro Gap functions and penalization for solving equilibrium problems with nonlinear constraints December 4, 2014 UnipiEprints view Download (402Kb) abstract The paper deals with equilibrium problems (EPs) with nonlinear convex constraints. First, EP is reformulated as a global optimization problem introducing a class of gap functions, in which the feasible set of EP is replaced by a polyhedral approximation. Then, an algorithm is given for solving EP through a descent type procedure related to exact penalties of the gap functions and its global convergence is proved. Finally, the algorithm is tested on a network oligopoly problem with nonlinear congestion constraints
Frangioni, Antonio and Gorgone, Enrico Generalized Bundle Methods for Sum-Functions with ''Easy'' Components: Applications to Multicommodity Network Design December 4, 2014 UnipiEprints view Download (281Kb) abstract We propose a modification to the (generalized) bundle scheme for minimization of a convex nondifferentiable sum-function in the case where some of the components are ''easy'', that is, they are Lagrangian functions of explicitly known convex programs with ''few'' variables and constraints. This happens in many practical cases, particularly within applications to combinatorial optimization. In this case one can insert in the master problem a suitably modified representation of the original convex subproblem, as an alternative to iteratively inner-approximating it by means of the produced extreme points, thus providing the algorithm with exact information about a part of the objective function, possibly leading to performance improvements. This is shown to happen in at least one relevant application: the computation of tight lower bounds for Fixed-Charge Multicommodity Min-Cost Flow problems. Buscemi, Maria Grazia and Coppo, Mario and Dezani-Ciancaglini, Mariangiola and Montanari, Ugo Constraints for Service Contracts December 3, 2014 UnipiEprints view Download (165Kb) abstract This paper focuses on client-service interactions distinguishing between three phases: negotiate, commit and execute. The participants negotiate their behaviours, and if an agreement is reached they commit and start an execution which is guaranteed to respect the interaction scheme agreed upon. These ideas are materialised through a calculus of contracts enriched with semiring-based constraints, which allow clients to choose services and to interact with them in a safe way. A concrete representation of these constraints with logic programs and logic program combinations is straightforward, thus reducing constraint solution (and consequently the establishment of a contract) to the execution of a logic program.

Bigi, Giancarlo and Frangioni, Antonio and Zhang, Qinghua Beyond canonical DC programs: the single reverse polar problem December 3, 2014 UnipiEprints view Download (363Kb) abstract We propose a novel generalization of the canonical DC problem (CDC), and we study the convergence of outer approximation algorithms for its solution which use an approximated oracle for checking the global optimality conditions. Although the approximated optimality conditions are similar to those of CDC, this new class of problems is shown to significantly differ from its special case. Indeed, outer approximation approaches for CDC need be substantially modified in order to cope with the more general problem, bringing to new algorithms. We develop a hierarchy of conditions that guarantee global convergence, and we build three different cutting plane algorithms relying on them.
Vanneschi, Marco and Mencagli, Gabriele QoS-control of Structured Parallel Computations: a Predictive Control Approach December 3, 2014 UnipiEprints view Download (732Kb) abstract A central issue for parallel applications executed on heterogeneous distributed platforms (e.g. Grids and Clouds) is assuring that performance and cost parameters are optimized throughout the execution. A solution is based on providing application components with adaptation strategies able to select at run-time the best component configuration. In this report we will introduce a preliminary work concerning the exploitation of control-theoretic techniques for controlling the Quality of Service of parallel computations. In particular we will demonstrate how the model-based predictive control strategy can be used based on first-principle performance models of structured parallelism schemes. We will also evaluate the viability of our approach on a first experimental scenario.
Bellia, Marco and Occhiuto, Maria Eugena Proving Translation and Reduction Semantics Equivalent for Java Simple Closures: Extended Version December 3, 2014 UnipiEprints view Download (317Kb) abstract \Lname\ is a minimal core calculus that extends Featherweight (generic) Java, \FGJ, with lambda expressions. It has been used to study properties of Simple Closure in Java, including type safety and the abstraction property. Its formalization is based on a reduction semantics and a typing system that extend those of \FGJ. ${\cal F}$ is a source-to-source, translation rule system from Java 1.5 extended with lambda expressions back to ordinary Java 1.5. It has been introduced to study implementation features of closures in Java, including assignment of non local variables and relations with anonymous class objects. In this paper we prove that the two semantics commute.

Romei, Andrea and Ruggieri, Salvatore A multidisciplinary survey on discrimination analysis December 3, 2014 UnipiEprints view Download (527Kb) abstract Most of the decisions in the today's knowledge society are taken
on the basis of historical data by extracting models, patterns,
profiles, and rules of human behavior in support of (automated)
decision making. There is then the need of developing models,
methods and technologies for modelling the processes of
discrimination analysis in order to discover and prevent
discrimination phenomena. In this respect, discrimination analysis
from data should build over the large body of existing legal and
economic studies. This paper intends to provide a
multi-disciplinary survey of the literature on discrimination data
analysis, including methods for data collection, empirical
studies, controlled experiments, statistical evidence, and their
legal requirements and grounds. We cover the following mainstream
research lines: labour economic models, (quasi-)experimental
approaches such as auditing and controlled experiments,
profiling-based approaches such as racial profiling and credit
markets, and the recently blooming research on knowledge discovery
approaches. Degano, Pierpaolo and Mezzetti, Gianluca and Ferrari, Gianluigi Nominal models and resource usage control December 3, 2014 UnipiEprints view Download (588Kb) abstract Two classes of nominal automata, namely Usage Automata (UAs) and
Variable Finite Automata (VFAs) are considered to express resource con-
trol policies over program execution traces expressed by a nominal calcu-
lus (Usages). We ?rst analyse closure properties of UAs, and then show
UAs less expressive than VFAs. We ?nally carry over VFAs the symbolic
technique for model checking Usages against UAs, so making it possi-
ble to verify the compliance of a program with a larger class of security
properties. Romei, Andrea and Turini, Franco Programming the KDD process using XQuery December 3, 2014 UnipiEprints view Download (286Kb) abstract XQuake is a language and system for programming data mining processes over native XML databases in the spirit of inductive databases. It extends XQuery to support KDD tasks. This paper focuses on the features required in the definition of the steps of the mining process. The main objective is to show the expressiveness of the language in handling mining operations as an extension of basic XQuery expressions. To this purpose, the paper offers an extended application in the field of analyzing web logs.
Mencagli, Gabriele and Vanneschi, Marco Performance Analysis of Computation Graphs of Parallel Modules December 3, 2014 UnipiEprints view Download (762Kb) abstract In this report we provide the fundamental results for applying a formal performance modeling of distributed parallel computations described as computation graphs of parallel modules. In our approach parallelism is expressed inside each module (based on structured parallelism schemes) and between different modules that can compose general graph structures. Our methodological approach, based on Queueing Theory and Queueing Network Theory foundations, provides the necessary tools for predicting the steady-state behavior of a parallel computation.
Marco, Bellia and Occhiuto, Maria Eugena {Java$\Omega$: Higher Order Programming in Java -- The Technical Complements December 3, 2014 UnipiEprints view Download (358Kb) abstract The paper contains the technical complements of the paper, of the same authors, published in the book "Java in Academia and Research", iConcept Press Ltd, 2011, that describes the extensions of Java 1.5 with Higher Order (H.O. for short) mechanisms. The present complements contain the complete tables of (i) The translation semantics of a form of closure for Java; (ii) The translation semantics of H.O. methods with mc\_parameters as arguments; (iii) The translation semantics of Java extended with the above mechanisms put together; (iv) Theorems, lemmas and proofs of all the properties of the mechanisms and of the semantics discussed in the paper.
Aldinucci, Marco and Anardu, L and Danelutto, Marco and Kilpatrick, Peter and Torquati, Massimo Targeting multi cores by structured programming and data flow December 3, 2014 UnipiEprints view Download (389Kb) abstract Data flow techniques have been around since the early ’70s when they were used in compilers for sequential languages. Shortly after their intro- duction they were also considered as a possible model for parallel comput- ing, although the impact here was limited. Recently, however, data flow has been identified as a candidate for efficient implementation of various programming models on multi-core architectures. In most cases, however, the burden of determining data flow “macro” instructions is left to the programmer, while the compiler/run time system manages only the ef- ficient scheduling of these instructions. We discuss a structured parallel programming approach supporting automatic compilation of programs to macro data flow and we show experimental results demonstrating the fea- sibility of the approach and the efficiency of the resulting “object” code on different classes of state-of-the-art multi-core architectures. The ex- perimental results use different base mechanisms to implement the macro data flow run time support, from plain pthreads with condition variables to more modern and effective lock- and fence-free parallel frameworks. Experimental results comparing efficiency of the proposed approach with those achieved using other, more classical parallel frameworks are also presented.

NOTA. Da gennaio 2015 il Technical reports possono essere inseriti nel deposito Open Access UnipiEprints.
Una volta pubblicati saranno visibili anche in queste pagine. L'importazione dei dati da UnipiEprints è ancora da perfezionare. Il vecchio sistema TR non è più mantenuto.

Dati importati da UnipiEprints