Time filter

Source Type

DAGAND P.-E.,CNRS Laboratory for Informatics
Journal of Functional Programming | Year: 2017

Functional programmers from all horizons strive to use, and sometimes abuse, their favorite type system in order to capture the invariants of their programs. A widely used tool in that trade consists in defining finely indexed datatypes. Operationally, these types classify the programmer's data, following the ML tradition. Logically, these types enforce the program invariants in a novel manner. This new programming pattern, by which one programs over inductive definitions to account for some invariants, lead to the development of a theory of ornaments (McBride, 2011 Ornamental Algebras, Algebraic Ornaments. Unpublished). However, ornaments originate as a dependently-typed object and may thus appear rather daunting to a functional programmer of the non-dependent kind. This article aims at presenting ornaments from first-principles and, in particular, to declutter their presentation from syntactic considerations. To do so, we shall give a sufficiently abstract model of indexed datatypes by means of many-sorted signatures. In this process, we formalize our intuition that an indexed datatype is the combination of a data-structure and a data-logic. Over this abstraction of datatypes, we shall recast the definition of ornaments, effectively giving a model of ornaments. Benefiting both from the operational and abstract nature of many-sorted signatures, ornaments should appear applicable and, one hopes, of interest beyond the type-theoretic circles, case in point being languages with generalized abstract datatypes or refinement types. Copyright © Cambridge University Press 2017

Beynier A.,CNRS Laboratory for Informatics
Proceedings - 2016 IEEE 28th International Conference on Tools with Artificial Intelligence, ICTAI 2016 | Year: 2016

Multiagent patrolling in adversarial domains has been widely studied in recent years. However, little attention has been paid to cooperation issues between patrolling agents. Moreover, most existing works focus on one-shot attacks and assume full rationality of the adversaries. Nonetheless, when patrolling frontiers, detecting illegal fishing or poaching; security forces face several adversaries with limited observability and rationality, that perform multiple illegal actions spread in time and space. In this paper, we develop a cooperative approach to improve defenders efficiency in such settings. We propose a new formalization of multiagent patrolling problems allowing for effective cooperation between the defenders. Our work accounts for uncertainty on action outcomes and partial observability of the system. Unlike existing security games, a generic model of the opponents is considered thus handling limited observability and bounded rationality of the adversaries. We then describe a learning mechanism allowing the defenders to take advantage of their observations about the adversaries and to compute cooperative patrolling strategies consequently. © 2016 IEEE.

Lauter C.,CNRS Laboratory for Informatics
Conference Record - Asilomar Conference on Signals, Systems and Computers | Year: 2016

Systems support mathematical functions like exp, sin, cos through mathematical libraries (libm). With increasingly parallel hardware, scalar libm functions do not suffice; implementations that work on vectors in an element-by-element (SIMD) fashion are required. Only few Open-Source implementations of vector libms exist. They are mostly written in assembly, which hinders portability and maintenance. As they depend on the scalar system libm for special case handling, the existing vector libms may induce a source of non-reproducibility. We present an Open-Source vector libm implemented with high-level scalar C that a modern compiler can translate to SIMD code. The error of all functions does not exceed 8ulp, while a performance gain of up to 278.5% is obtained. Our library is fully free-standing, i.e. it does not depend on any other system library. © 2016 IEEE.

El Din M.S.,CNRS Laboratory for Informatics | Zhi L.,Key Laboratory of Mathematics Mechanization
SIAM Journal on Optimization | Year: 2010

Let P = {h1, . . . , hs} ⊂ Z[Y1, . . . ,Yk], D ≥ deg(hi) for 1 ≤ i ≤ s, ς bounding the bit length of the coefficients of the hi's, and let ? be a quantifier-free P-formula defining a convex semialgebraic set. We design an algorithm returning a rational point in S if and only if S∩Q ≠ ∅. It requires ςO(1)DO(k3) bit operations. If a rational point is outputted, its coordinates have bit length dominated by ςDO(k3). Using this result, we obtain a procedure for deciding whether a polynomial f ∈ Z[X1, . . . ,Xn] is a sum of squares of polynomials in Q[X1, . . . ,Xn]. Denote by d the degree of f, ? the maximum bit length of the coefficients in f, D = (n+d/n), and k ≤ D(D + 1) ?(n+d/n), This procedure requires τO(1)DO(k3) bit operations, and the coefficients of the outputted polynomials have bit length dominated by τDO(k3). © 2010 Society for Industrial and Applied Mathematics.

Escoffier B.,CNRS Laboratory for Informatics
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2016

Max coloring is a well known generalization of the usual Min Coloring problem, widely studied from (standard) complexity and approximation viewpoints. Here, we tackle this problem under the framework of parameterized complexity. In particular, we first show to what extend the result of [3] - saving colors from the trivial bound of n on the chromatic number - extends to Max Coloring. Then we consider possible improvements of these results by considering the problem of saving colors/weight with respect to a better bound on the chromatic number. Finally, we consider the fixed parameterized tractability of Max Coloring in restricted graph classes under standard parameterization. © Springer-Verlag GmbH Germany 2016.

Viappiani P.,CNRS Laboratory for Informatics
IJCAI International Joint Conference on Artificial Intelligence | Year: 2015

Positional scoring rules are often used for rank aggregation. In this work we study how scoring rules can be formulated as the minimization of some distance measures between rankings, and we also consider a new family of aggregation methods, called biased scoring rules. This work extends a previous known observation connecting Borda count with the minimization of the sum of the Spearman distances (calculated with respect to a set of input rankings). In particular we consider generalizations of the Spearman distance that can give different weights to items and positions; we also handle the case of incomplete rank data. This has applications in the clustering of rank data, where two main steps need to be performed: aggregating rankings of the same cluster into a representative ranking (the cluster's centroid) and assigning each ranking to its closest centroid. Using the proper combination of scoring rules (for aggregation) and distances (for assignment), it is possible to perform clustering in a computationally efficient way and as well account for specific desired behaviors (give more weight to top positions, bias the centroids in favor of particular items).

Angelopoulos S.,CNRS Laboratory for Informatics
IJCAI International Joint Conference on Artificial Intelligence | Year: 2015

This paper addresses two classes of different, yet interrelated optimization problems. The first class of problems involves a robot that must locate a hidden target in an environment that consists of a set of concurrent rays. The second class pertains to the design of interruptible algorithms by means of a schedule of contract algorithms. We study several variants of these families of problems, such as searching and scheduling with probabilistic considerations, redundancy and fault-tolerance issues, randomized strategies, and trade-offs between performance and preemptions. For many of these problems we present the first known results that apply to multi-ray and multi-problem domains. Our objective is to demonstrate that several wellmotivated settings can be addressed using a common approach.

Labroche N.,CNRS Laboratory for Informatics
Neurocomputing | Year: 2014

This paper describes two new online fuzzy clustering algorithms based on medoids. These algorithms have been developed to deal with either very large datasets that do not fit in main memory or data streams in which data are produced continuously. The innovative aspect of our approach is the combination of fuzzy methods, which are well adapted to outliers and overlapping clusters, with medoids and the introduction of a decay mechanism to adapt more effectively to changes over time in the data streams. The use of medoids instead of means allows to deal with non-numerical data (e.g. sequences. .) and improves the interpretability of the cluster centers. Experiments conducted on artificial and real datasets show that our new algorithms are competitive with state-of-the-art clustering algorithms in terms of purity of the partition, F1 score and computation times. Finally, experiments conducted on artificial data streams show the benefit of our decay mechanism in the case of evolving distributions. © 2013 Elsevier B.V.

Lamaury J.,CNRS Laboratory for Informatics | Gouttefarde M.,CNRS Laboratory for Informatics
Proceedings - IEEE International Conference on Robotics and Automation | Year: 2013

This paper deals with the control of a 6-DOF cable-suspended parallel robot (CSPR) able to perform tasks such as pick-and-place trajectories over a large workspace. In order to maximize the ratio between the robot workspace and its overall dimensions, actuation redundancy can be used. The control of such a redundantly actuated CSPR turns to be challenging as a realtime embeddable algorithm for distributing the cable tensions should be used, together with a suitable control scheme. This paper proposes a computationally efficient tension distribution algorithm implemented within a dual-space feedforward scheme in order to properly control the moving platform. Experimentations are performed on a large 6-DOF CSPR prototype equipped with 8 actuators. © 2013 IEEE.

Cornaz D.,CNRS Laboratory for the Analysis and Modeling of Decision Systems | Galand L.,CNRS Laboratory for the Analysis and Modeling of Decision Systems | Spanjaard O.,CNRS Laboratory for Informatics
IJCAI International Joint Conference on Artificial Intelligence | Year: 2013

This paper is devoted to complexity results regarding specific measures of proximity to single-peakedness and single-crossingness, called "single-peaked width" [Cornaz et al., 2012] and "single-crossing width". Thanks to the use of the PQ-tree data structure [Booth and Lueker, 1976], we show that both problems are polynomial time solvable in the general case (while it was only known for single-peaked width and in the case of narcissistic preferences). Furthermore, we establish one of the first results (to our knowledge) concerning the effect of nearly single-peaked electorates on the complexity of an NP-hard voting system, namely we show the fixed-parameter tractability of Kemeny elections with respect to the parameters "single-peaked width" and "single-crossing width".

Loading CNRS Laboratory for Informatics collaborators
Loading CNRS Laboratory for Informatics collaborators