This introductory text to statistical machine translation (SMT) provides all of the theories and methods needed to build a statistical machine translator, such as Google Language Tools and Babelfish. In general, statistical techniques allow automatic translation systems to be built quickly for any language-pair using only translated texts and generic software. With increasing globalization, statistical machine translation will be central to communication and commerce. Based on courses and tutorials, and classroom-tested globally, it is ideal for instruction or self-study, for advanced undergraduates and graduate students in computer science and/or computational linguistics, and researchers in natural language processing. The companion website provides open-source corpora and tool-kits.

/pdf/statistical-machine-translation-1yr1jtzcjp.pdf

Statistical Machine Translation

A newly deployed multi-hop radio network is unstructured and lacks a reliable and efficient communication scheme. In this paper, we take a step towards analyzing the problems existing during the initialization phase of ad hoc and sensor networks. Particularly, we model the network as a multi-hop quasi unit disk graph and allow nodes to wake up asynchronously at any time. Further, nodes do not feature a reliable collision detection mechanism, and they have only limited knowledge about the network topology. We show that even for this restricted model, a good clustering can be computed efficiently. Our algorithm efficiently computes an asymptotically optimal clustering. Based on this algorithm, we describe a protocol for quickly establishing synchronized sleep and listen schedule between nodes within a cluster. Additionally, we provide simulation results in a variety of settings.

/pdf/initializing-newly-deployed-ad-hoc-and-sensor-networks-2t97y6n9b7.pdf

Initializing newly deployed ad hoc and sensor networks

Cette derniere Newsletter avant les Grandes Vacances est, pur concours de circonstances, entierement consacree a l’Education. Mais l’Education sous l’angle de son importance pour la consecration dans l’excellence. La consecration, c’est ce qu’a obtenu Gisele Mophou-Massengo, quand son jury de soutenance l’a declaree Habilitee a Diriger des Recherches. C’est egalement la reconnaissance qui est accordee a Elina Marie Devoue par le jury du prix « Religions et Developpement » pour son article consacre aux Religions et a leurs impacts sur la societe dans la Caraibe.La reconnaissance du savoir et du savoir-faire ne passe pas forcement par sa consecration. Elle passe egalement par sa diffusion la plus large. C’est ce que tente de faire le Ceregmia en organisant depuis 2009 ses « seminaires intra Ceregmia » qui permettent l’expression de la transversalite du laboratoire sur des chantiers en cours ou deja finalises autour de thematiques que nous vous invitons a decouvrir en page 2. De meme, le Ceregmia a lance une nouvelle publication en 2010, ses « Documents de Travail ». Ces Documents d’audience plus confidentielle sont reellement reserves a un public de scientifiques. Diffuser le savoir en l’adaptant aux evolutions de l’environnement et aux exigences du marche. C’est ce qu’a fait le Ceregmia grâce a sa nouvelle offre de formation en Masters qui a ete habilitee par le Ministere et qui ouvrira des la rentree 2010 (p. 4). Par ailleurs, nous avions eu l’occasion de vous entretenir des changements intervenus a l’IFGCar suite au seisme du 12 janvier 2010 en Haiti. Nous sommes heureux aujourd’hui, de vous annoncer que l’IFGCar qui est desormais l’IAC retourne s’installer en Haiti. Les formations en Master reprendront des le 1 er . Octobre. Enfin, Nous vous signalons la parution de deux ouvrages aux Editions Ibis Rouge ils vous invitent a decouvrir la Guyane sous un angle moins connu, celui de l’anthropologie. Charley Granvorka

/pdf/habilitation-a-diriger-des-recherches-1hfqqvwsee.pdf

Habilitation a diriger des recherches

Distributed data aggregation is an important task, allowing the decentralized determination of meaningful global properties, which can then be used to direct the execution of other applications. The resulting values are derived by the distributed computation of functions like Count , Sum , and Average . Some application examples deal with the determination of the network size, total storage capacity, average load, majorities and many others. In the last decade, many different approaches have been proposed, with different trade-offs in terms of accuracy, reliability, message and time complexity. Due to the considerable amount and variety of aggregation algorithms, it can be difficult and time consuming to determine which techniques will be more appropriate to use in specific settings, justifying the existence of a survey to aid in this task. This work reviews the state of the art on distributed data aggregation algorithms, providing three main contributions. First, it formally defines the concept of aggregation, characterizing the different types of aggregation functions. Second, it succinctly describes the main aggregation techniques, organizing them in a taxonomy. Finally, it provides some guidelines toward the selection and use of the most relevant techniques, summarizing their principal characteristics.

/pdf/a-survey-of-distributed-data-aggregation-algorithms-16eiyu3imz.pdf

A Survey of Distributed Data Aggregation Algorithms

ion-Driven SAT-based Analysis of Security Protocols . . . . . . . . . . 257 Alessandro Armando, Luca Compagna A Case for Efficient Solution Enumeration . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 Sarfraz Khurshid, Darko Marinov, Ilya Shlyakhter, Daniel Jackson Cache Performance of SAT Solvers: a Case Study for Efficient Implementation of Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 Lintao Zhang, Sharad Malik Local Consistencies in SAT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 Christian Bessière, Emmanuel Hebrard, Toby Walsh Guiding SAT Diagnosis with Tree Decompositions . . . . . . . . . . . . . . . . . . . . 315 Per Bjesse, James Kukula, Robert Damiano, Ted Stanion, Yunshan Zhu On Computing k-CNF Formula Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 330 Ryan Williams Effective Preprocessing with Hyper-Resolution and Equality Reduction . . 341 Fahiem Bacchus, Jonathan Winter Read-Once Unit Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356 Hans Kleine Büning, Xishun Zhao The Interaction Between Inference and Branching Heuristics . . . . . . . . . . . . 370 Lyndon Drake, Alan Frisch Hypergraph Reductions and Satisfiability Problems . . . . . . . . . . . . . . . . . . . 383 Daniele Pretolani Table of

Theory and Applications of Satisfiability Testing

We consider the problem of waking up n processors of a completely broadcast system. We analyze this problem in both globally and locally synchronous models, with or without n being known to processors and with or without labeling of processors. The main question we answer is: how fast we can wake all the processors up with probability 1 - ? in each of these eight models. In [10] a logarithmic waking algorithm for the strongest set of assumptions is described, while for weaker models only linear and quadratic algorithms were obtained. We prove that in the weakest model (local synchronization, no knowledge of n or labeling) the best waking time is O(n/ log n). We also show logarithmic or poly-logarithmic waking algorithms for all stronger models, which in some cases gives an exponential improvement over previous results.

/pdf/probabilistic-algorithms-for-the-wakeup-problem-in-single-1ys993cqwk.pdf

Probabilistic Algorithms for the Wakeup Problem in Single-Hop Radio Networks

We present energy efficient algorithms for leader election in single channel single-hop radio networks with no collision detection. We present a deterministic solution with sublogarithmic energy cost (the best previous result was O(log n)) and show a double logarithmic lower bound. We prove that this lower bound holds in a randomized case, in a certain sense.For the case, when the number n of active stations can be approximated in advance, we show a randomized algorithm with energy consumption O(log* n) that yields a result with high probability (the best previous result was O(log log n)).

Efficient algorithms for leader election in radio networks

Algorithms for radio networks are studied in two scenarios: (a) the number of active stations is known (or approximately known) (b) the number of active stations is unknown. In the second (more realistic) case it is much harder to design efficient algorithms. For this reason, we design an efficient randomized algorithm for a single-hop radio network that approximately counts the number of its active stations. With probability higher than 1 - 1/n, this approximation is within a constant factor, the algorithm runs in poly-logarithmic time and its energy cost is o(log log n). This improves the previous O(log n) bound for energy. In particular, our algorithm can be applied to improve energy cost of known leader election and initialization protocols (without loss of time efficiency).

Energy-Efficient Size Approximation of Radio Networks with No Collision Detection

We present a distributed randomized algorithm finding Minimum Spanning Tree (MST) of a given graph in O(1) rounds, with high probability, in the congested clique model. The input graph in the congested clique model is a graph of n nodes, where each node initially knows only its incident edges. The communication graph is a clique with limited edge bandwidth: each two nodes (not necessarily neighbours in the input graph) can exchange O(log n) bits. As in previous works, the key part of the MST algorithm is an efficient Connected Components (CC) algorithm. However, unlike the former approaches, we do not aim at simulating the standard Boruvka's algorithm, at least at initial stages of the CC algorithm. Instead, we develop a new technique which combines connected components of sample sparse subgraphs of the input graph in order to accelerate the process of uncovering connected components of the original input graph. More specifically, we develop a sparsification technique which reduces an initial CC problem in O(1) rounds to its two restricted instances. The former instance has a graph with maximal degree O(log log n) as the input - here our sample-combining technique helps. In the latter instance, a partition of the input graph into O(n/ log log n) connected components is known. This gives an opportunity to apply previous algorithms to determine connected components in O(1) rounds. Our result addresses a problem proposed by Lotker et al. [SPAA 2003; SICOMP 2005] and improves over previous O(log* n) algorithm of Ghaffari et al. [PODC 2016], and O(log log log n) algorithm of Hegeman et al. [PODC 2015]. It also determines Θ(1) round complexity in the congested clique for MST, as well as other graph problems, including bipartiteness, cut verification, s-t connectivity, and cycle containment.

MST in O(1) rounds of congested clique

We present a distributed randomized algorithm finding Minimum Spanning Tree (MST) of a given graph in O(1) rounds, with high probability, in the Congested Clique model. The input graph in the Congested Clique model is a graph of n nodes, where each node initially knows only its incident edges. The communication graph is a clique with limited edge bandwidth: each two nodes (not necessarily neighbours in the input graph) can exchange $O(\log n)$ bits. 
As in previous works, the key part of the MST algorithm is an efficient Connected Components (CC) algorithm. However, unlike the former approaches, we do not aim at simulating the standard Boruvka algorithm, at least at initial stages of the CC algorithm. Instead, we develop a new technique which combines connected components of sample sparse subgraphs of the input graph in order to accelerate the process of uncovering connected components of the original input graph. More specifically, we develop a sparsification technique which reduces an initial CC problem in $O(1)$ rounds to its two restricted instances. The former instance has a graph with maximal degree $O(\log \log n)$ as the input -- here our sample-combining technique helps. In the latter instance, a partition of the input graph into $O(n/\log \log n)$ connected components is known. This gives an opportunity to apply previous algorithms to determine connected components in $O(1)$ rounds. 
Our result addresses the problem from and the $O(\log \log n)$ algorithm of Lotker et al. [SPAA 2003; SICOMP 2005], improves over previous $O(\log* n)$ algorithm of Ghaffari et al. [PODC 2016] and $O(\log \log \log n)$ algorithm of Hegeman et al. [PODC 2015] . It also determines $\Theta(1)$ round complexity in the congested clique for MST, as well as other graph problems, including bipartiteness, cut verification, s-t connectivity and cycle containment.

Tomasz Jurdzinski

Papers

Probabilistic Algorithms for the Wakeup Problem in Single-Hop Radio Networks

Efficient algorithms for leader election in radio networks

Energy-Efficient Size Approximation of Radio Networks with No Collision Detection

MST in O(1) rounds of congested clique

MST in O(1) Rounds of the Congested Clique