text
stringlengths 0
4.09k
|
|---|
Abstract: Pac-Bayes bounds are among the most accurate generalization bounds for classifiers learned from independently and identically distributed (IID) data, and it is particularly so for margin classifiers: there have been recent contributions showing how practical these bounds can be either to perform model selection (Ambroladze et al., 2007) or even to directly guide the learning of linear classifiers (Germain et al., 2009). However, there are many practical situations where the training data show some dependencies and where the traditional IID assumption does not hold. Stating generalization bounds for such frameworks is therefore of the utmost interest, both from theoretical and practical standpoints. In this work, we propose the first - to the best of our knowledge - Pac-Bayes generalization bounds for classifiers trained on data exhibiting interdependencies. The approach undertaken to establish our results is based on the decomposition of a so-called dependency graph that encodes the dependencies within the data, in sets of independent data, thanks to graph fractional covers. Our bounds are very general, since being able to find an upper bound on the fractional chromatic number of the dependency graph is sufficient to get new Pac-Bayes bounds for specific settings. We show how our results can be used to derive bounds for ranking statistics (such as Auc) and classifiers trained on data distributed according to a stationary \ss-mixing process. In the way, we show how our approach seemlessly allows us to deal with U-processes. As a side note, we also provide a Pac-Bayes generalization bound for classifiers learned on data from stationary $\varphi$-mixing distributions.
|
Title: Sparsity and `Something Else': An Approach to Encrypted Image Folding
|
Abstract: A property of sparse representations in relation to their capacity for information storage is discussed. It is shown that this feature can be used for an application that we term Encrypted Image Folding. The proposed procedure is realizable through any suitable transformation. In particular, in this paper we illustrate the approach by recourse to the Discrete Cosine Transform and a combination of redundant Cosine and Dirac dictionaries. The main advantage of the proposed technique is that both storage and encryption can be achieved simultaneously using simple processing steps.
|
Title: Paired Comparisons-based Interactive Differential Evolution
|
Abstract: We propose Interactive Differential Evolution (IDE) based on paired comparisons for reducing user fatigue and evaluate its convergence speed in comparison with Interactive Genetic Algorithms (IGA) and tournament IGA. User interface and convergence performance are two big keys for reducing Interactive Evolutionary Computation (IEC) user fatigue. Unlike IGA and conventional IDE, users of the proposed IDE and tournament IGA do not need to compare whole individuals each other but compare pairs of individuals, which largely decreases user fatigue. In this paper, we design a pseudo-IEC user and evaluate another factor, IEC convergence performance, using IEC simulators and show that our proposed IDE converges significantly faster than IGA and tournament IGA, i.e. our proposed one is superior to others from both user interface and convergence performance points of view.
|
Title: Approximate Nearest Neighbor Search through Comparisons
|
Abstract: This paper addresses the problem of finding the nearest neighbor (or one of the R-nearest neighbors) of a query object q in a database of n objects. In contrast with most existing approaches, we can only access the ``hidden'' space in which the objects live through a similarity oracle. The oracle, given two reference objects and a query object, returns the reference object closest to the query object. The oracle attempts to model the behavior of human users, capable of making statements about similarity, but not of assigning meaningful numerical values to distances between objects.
|
Title: Universal and Composite Hypothesis Testing via Mismatched Divergence
|
Abstract: For the universal hypothesis testing problem, where the goal is to decide between the known null hypothesis distribution and some other unknown distribution, Hoeffding proposed a universal test in the nineteen sixties. Hoeffding's universal test statistic can be written in terms of Kullback-Leibler (K-L) divergence between the empirical distribution of the observations and the null hypothesis distribution. In this paper a modification of Hoeffding's test is considered based on a relaxation of the K-L divergence test statistic, referred to as the mismatched divergence. The resulting mismatched test is shown to be a generalized likelihood-ratio test (GLRT) for the case where the alternate distribution lies in a parametric family of the distributions characterized by a finite dimensional parameter, i.e., it is a solution to the corresponding composite hypothesis testing problem. For certain choices of the alternate distribution, it is shown that both the Hoeffding test and the mismatched test have the same asymptotic performance in terms of error exponents. A consequence of this result is that the GLRT is optimal in differentiating a particular distribution from others in an exponential family. It is also shown that the mismatched test has a significant advantage over the Hoeffding test in terms of finite sample size performance. This advantage is due to the difference in the asymptotic variances of the two test statistics under the null hypothesis. In particular, the variance of the K-L divergence grows linearly with the alphabet size, making the test impractical for applications involving large alphabet distributions. The variance of the mismatched divergence on the other hand grows linearly with the dimension of the parameter space, and can hence be controlled through a prudent choice of the function class defining the mismatched divergence.
|
Title: Logic with Verbs
|
Abstract: The aim of this paper is to introduce a logic in which nouns and verbs are handled together as a deductive reasoning, and also to observe the relationship between nouns and verbs as well as between logics and conversations.
|
Title: A Nonconformity Approach to Model Selection for SVMs
|
Abstract: We investigate the issue of model selection and the use of the nonconformity (strangeness) measure in batch learning. Using the nonconformity measure we propose a new training algorithm that helps avoid the need for Cross-Validation or Leave-One-Out model selection strategies. We provide a new generalisation error bound using the notion of nonconformity to upper bound the loss of each test example and show that our proposed approach is comparable to standard model selection methods, but with theoretical guarantees of success and faster convergence. We demonstrate our novel model selection technique using the Support Vector Machine.
|
Title: Back analysis based on SOM-RST system
|
Abstract: This paper describes application of information granulation theory, on the back analysis of Jeffrey mine southeast wall Quebec. In this manner, using a combining of Self Organizing Map (SOM) and rough set theory (RST), crisp and rough granules are obtained. Balancing of crisp granules and sub rough granules is rendered in close-open iteration. Combining of hard and soft computing, namely finite difference method (FDM) and computational intelligence and taking in to account missing information are two main benefits of the proposed method. As a practical example, reverse analysis on the failure of the southeast wall Jeffrey mine is accomplished.
|
Title: Clustering Based on Pairwise Distances When the Data is of Mixed Dimensions
|
Abstract: In the context of clustering, we consider a generative model in a Euclidean ambient space with clusters of different shapes, dimensions, sizes and densities. In an asymptotic setting where the number of points becomes large, we obtain theoretical guaranties for a few emblematic methods based on pairwise distances: a simple algorithm based on the extraction of connected components in a neighborhood graph; the spectral clustering method of Ng, Jordan and Weiss; and hierarchical clustering with single linkage. The methods are shown to enjoy some near-optimal properties in terms of separation between clusters and robustness to outliers. The local scaling method of Zelnik-Manor and Perona is shown to lead to a near-optimal choice for the scale in the first two methods. We also provide a lower bound on the spectral gap to consistently choose the correct number of clusters in the spectral method.
|
Title: An Efficient Secure Multimodal Biometric Fusion Using Palmprint and Face Image
|
Abstract: Biometrics based personal identification is regarded as an effective method for automatically recognizing, with a high confidence a person's identity. A multimodal biometric systems consolidate the evidence presented by multiple biometric sources and typically better recognition performance compare to system based on a single biometric modality. This paper proposes an authentication method for a multimodal biometric system identification using two traits i.e. face and palmprint. The proposed system is designed for application where the training data contains a face and palmprint. Integrating the palmprint and face features increases robustness of the person authentication. The final decision is made by fusion at matching score level architecture in which features vectors are created independently for query measures and are then compared to the enrolment template, which are stored during database preparation. Multimodal biometric system is developed through fusion of face and palmprint recognition.
|
Title: Similarity Matching Techniques for Fault Diagnosis in Automotive Infotainment Electronics
|
Abstract: Fault diagnosis has become a very important area of research during the last decade due to the advancement of mechanical and electrical systems in industries. The automobile is a crucial field where fault diagnosis is given a special attention. Due to the increasing complexity and newly added features in vehicles, a comprehensive study has to be performed in order to achieve an appropriate diagnosis model. A diagnosis system is capable of identifying the faults of a system by investigating the observable effects (or symptoms). The system categorizes the fault into a diagnosis class and identifies a probable cause based on the supplied fault symptoms. Fault categorization and identification are done using similarity matching techniques. The development of diagnosis classes is done by making use of previous experience, knowledge or information within an application area. The necessary information used may come from several sources of knowledge, such as from system analysis. In this paper similarity matching techniques for fault diagnosis in automotive infotainment applications are discussed.
|
Title: Performing Hybrid Recommendation in Intermodal Transportation-the FTMarket System's Recommendation Module
|
Abstract: Diverse recommendation techniques have been already proposed and encapsulated into several e-business applications, aiming to perform a more accurate evaluation of the existing information and accordingly augment the assistance provided to the users involved. This paper reports on the development and integration of a recommendation module in an agent-based transportation transactions management system. The module is built according to a novel hybrid recommendation technique, which combines the advantages of collaborative filtering and knowledge-based approaches. The proposed technique and supporting module assist customers in considering in detail alternative transportation transactions that satisfy their requests, as well as in evaluating completed transactions. The related services are invoked through a software agent that constructs the appropriate knowledge rules and performs a synthesis of the recommendation policy.
|
Title: Implementation of Rule Based Algorithm for Sandhi-Vicheda Of Compound Hindi Words
|
Abstract: Sandhi means to join two or more words to coin new word. Sandhi literally means `putting together' or combining (of sounds), It denotes all combinatory sound-changes effected (spontaneously) for ease of pronunciation. Sandhi-vicheda describes [5] the process by which one letter (whether single or cojoined) is broken to form two words. Part of the broken letter remains as the last letter of the first word and part of the letter forms the first letter of the next letter. Sandhi- Vicheda is an easy and interesting way that can give entirely new dimension that add new way to traditional approach to Hindi Teaching. In this paper using the Rule based algorithm we have reported an accuracy of 60-80% depending upon the number of rules to be implemented.
|
Title: Design of an ultrasound-guided robotic brachytherapy needle insertion system
|
Abstract: In this paper we describe a new robotic brachytherapy needle-insertion system that is designed to replace the template used in the manual technique. After a brief review of existing robotic systems, we describe the requirements that we based our design upon. A detailed description of the proposed system follows. Our design is capable of positioning and inclining a needle within the same workspace as the manual template. To help improve accuracy, the needle can be rotated about its axis during insertion into the prostate. The system can be mounted on existing steppers and also easily accommodates existing seed dispensers, such as the Mick Applicator.
|
Title: Stochastic Optimization of Linear Dynamic Systems with Parametric Uncertainties
|
Abstract: This paper describes a new approach to solving some stochastic optimization problems for linear dynamic system with various parametric uncertainties. Proposed approach is based on application of tensor formalism for creation the mathematical model of parametric uncertainties. Within proposed approach following problems are considered: prediction, data processing and optimal control. Outcomes of carried out simulation are used as illustration of properties and effectiveness of proposed methods.
|
Title: Reference Resolution within the Framework of Cognitive Grammar
|
Abstract: Following the principles of Cognitive Grammar, we concentrate on a model for reference resolution that attempts to overcome the difficulties previous approaches, based on the fundamental assumption that all reference (independent on the type of the referring expression) is accomplished via access to and restructuring of domains of reference rather than by direct linkage to the entities themselves. The model accounts for entities not explicitly mentioned but understood in a discourse, and enables exploitation of discursive and perceptual context to limit the set of potential referents for a given referring expression. As the most important feature, we note that a single mechanism is required to handle what are typically treated as diverse phenomena. Our approach, then, provides a fresh perspective on the relations between Cognitive Grammar and the problem of reference.
|
Title: Marking-up multiple views of a Text: Discourse and Reference
|
Abstract: We describe an encoding scheme for discourse structure and reference, based on the TEI Guidelines and the recommendations of the Corpus Encoding Specification (CES). A central feature of the scheme is a CES-based data architecture enabling the encoding of and access to multiple views of a marked-up document. We describe a tool architecture that supports the encoding scheme, and then show how we have used the encoding scheme and the tools to perform a discourse analytic task in support of a model of global discourse cohesion called Veins Theory (Cristea & Ide, 1998).
|
Title: A Common XML-based Framework for Syntactic Annotations
|
Abstract: It is widely recognized that the proliferation of annotation schemes runs counter to the need to re-use language resources, and that standards for linguistic annotation are becoming increasingly mandatory. To answer this need, we have developed a framework comprised of an abstract model for a variety of different annotation types (e.g., morpho-syntactic tagging, syntactic annotation, co-reference annotation, etc.), which can be instantiated in different ways depending on the annotator's approach and goals. In this paper we provide an overview of the framework, demonstrate its applicability to syntactic annotation, and show how it can contribute to comparative evaluation of parser output and diverse syntactic annotation schemes.
|
Title: Standards for Language Resources
|
Abstract: This paper presents an abstract data model for linguistic annotations and its implementation using XML, RDF and related standards; and to outline the work of a newly formed committee of the International Standards Organization (ISO), ISO/TC 37/SC 4 Language Resource Management, which will use this work as its starting point. The primary motive for presenting the latter is to solicit the participation of members of the research community to contribute to the work of the committee.
|
Title: Computing p-values of LiNGAM outputs via Multiscale Bootstrap
|
Abstract: Structural equation models and Bayesian networks have been widely used to study causal relationships between continuous variables. Recently, a non-Gaussian method called LiNGAM was proposed to discover such causal models and has been extended in various directions. An important problem with LiNGAM is that the results are affected by the random sampling of the data as with any statistical method. Thus, some analysis of the statistical reliability or confidence level should be conducted. A common method to evaluate a confidence level is a bootstrap method. However, a confidence level computed by ordinary bootstrap method is known to be biased as a probability-value ($p$-value) of hypothesis testing. In this paper, we propose a new procedure to apply an advanced bootstrap method called multiscale bootstrap to compute confidence levels, i.e., p-values, of LiNGAM outputs. The multiscale bootstrap method gives unbiased $p$-values with asymptotic much higher accuracy. Experiments on artificial data demonstrate the utility of our approach.
|
Title: Distribution-Specific Agnostic Boosting
|
Abstract: We consider the problem of boosting the accuracy of weak learning algorithms in the agnostic learning framework of Haussler (1992) and Kearns et al. (1992). Known algorithms for this problem (Ben-David et al., 2001; Gavinsky, 2002; Kalai et al., 2008) follow the same strategy as boosting algorithms in the PAC model: the weak learner is executed on the same target function but over different distributions on the domain. We demonstrate boosting algorithms for the agnostic learning framework that only modify the distribution on the labels of the points (or, equivalently, modify the target function). This allows boosting a distribution-specific weak agnostic learner to a strong agnostic learner with respect to the same distribution. When applied to the weak agnostic parity learning algorithm of Goldreich and Levin (1989) our algorithm yields a simple PAC learning algorithm for DNF and an agnostic learning algorithm for decision trees over the uniform distribution using membership queries. These results substantially simplify Jackson's famous DNF learning algorithm (1994) and the recent result of Gopalan et al. (2008). We also strengthen the connection to hard-core set constructions discovered by Klivans and Servedio (1999) by demonstrating that hard-core set constructions that achieve the optimal hard-core set size (given by Holenstein (2005) and Barak et al. (2009)) imply distribution-specific agnostic boosting algorithms. Conversely, our boosting algorithm gives a simple hard-core set construction with an (almost) optimal hard-core set size.
|
Title: A Convergent Online Single Time Scale Actor Critic Algorithm
|
Abstract: Actor-Critic based approaches were among the first to address reinforcement learning in a general setting. Recently, these algorithms have gained renewed interest due to their generality, good convergence properties, and possible biological relevance. In this paper, we introduce an online temporal difference based actor-critic algorithm which is proved to converge to a neighborhood of a local maximum of the average reward. Linear function approximation is used by the critic in order estimate the value function, and the temporal difference signal, which is passed from the critic to the actor. The main distinguishing feature of the present convergence proof is that both the actor and the critic operate on a similar time scale, while in most current convergence proofs they are required to have very different time scales in order to converge. Moreover, the same temporal difference signal is used to update the parameters of both the actor and the critic. A limitation of the proposed approach, compared to results available for two time scale convergence, is that convergence is guaranteed only to a neighborhood of an optimal value, rather to an optimal value itself. The single time scale and identical temporal difference signal used by the actor and the critic, may provide a step towards constructing more biologically realistic models of reinforcement learning in the brain.
|
Title: Language Models for Handwritten Short Message Services
|
Abstract: Handwriting is an alternative method for entering texts composing Short Message Services. However, a whole new language features the texts which are produced. They include for instance abbreviations and other consonantal writing which sprung up for time saving and fashion. We have collected and processed a significant number of such handwriting SMS, and used various strategies to tackle this challenging area of handwriting recognition. We proposed to study more specifically three different phenomena: consonant skeleton, rebus, and phonetic writing. For each of them, we compare the rough results produced by a standard recognition system with those obtained when using a specific language model.
|
Title: Vers la reconnaissance de mini-messages manuscrits
|
Abstract: Handwriting is an alternative method for entering texts which composed Short Message Services. However, a whole new language features the texts which are produced. They include for instance abbreviations and other consonantal writing which sprung up for time saving and fashion. We have collected and processed a significant number of such handwritten SMS, and used various strategies to tackle this challenging area of handwriting recognition. We proposed to study more specifically three different phenomena: consonant skeleton, rebus, and phonetic writing. For each of them, we compare the rough results produced by a standard recognition system with those obtained when using a specific language model to take care of them.
|
Title: Spatial Clustering Tests Based on Domination Number of a New Random Digraph Family
|
Abstract: We use the domination number of a parametrized random digraph family called proportional-edge proximity catch digraphs (PCDs) for testing multivariate spatial point patterns. This digraph family is based on relative positions of data points from various classes. We extend the results on the distribution of the domination number of proportional-edge PCDs, and use the domination number as a statistic for testing segregation and association against complete spatial randomness. We demonstrate that the domination number of the PCD has binomial distribution when size of one class is fixed while the size of the other (whose points constitute the vertices of the digraph) tends to infinity and asymptotic normality when sizes of both classes tend to infinity. We evaluate the finite sample performance of the test by Monte Carlo simulations, prove the consistency of the test under the alternatives, and suggest corrections for the support restriction on the class of points of interest and for small samples. We find the optimal parameters for testing each of the segregation and association alternatives. Furthermore, the methodology discussed in this article is valid for data in higher dimensions also.
|
Title: Cross-Validation for Unsupervised Learning
|
Abstract: Cross-validation (CV) is a popular method for model-selection. Unfortunately, it is not immediately obvious how to apply CV to unsupervised or exploratory contexts. This thesis discusses some extensions of cross-validation to unsupervised learning, specifically focusing on the problem of choosing how many principal components to keep. We introduce the latent factor model, define an objective criterion, and show how CV can be used to estimate the intrinsic dimensionality of a data set. Through both simulation and theory, we demonstrate that cross-validation is a valuable tool for unsupervised learning.
|
Title: Median K-flats for hybrid linear modeling with many outliers
|
Abstract: We describe the Median K-Flats (MKF) algorithm, a simple online method for hybrid linear modeling, i.e., for approximating data by a mixture of flats. This algorithm simultaneously partitions the data into clusters while finding their corresponding best approximating l1 d-flats, so that the cumulative l1 error is minimized. The current implementation restricts d-flats to be d-dimensional linear subspaces. It requires a negligible amount of storage, and its complexity, when modeling data consisting of N points in D-dimensional Euclidean space with K d-dimensional linear subspaces, is of order O(n K d D+n d^2 D), where n is the number of iterations required for convergence (empirically on the order of 10^4). Since it is an online algorithm, data can be supplied to it incrementally and it can incrementally produce the corresponding output. The performance of the algorithm is carefully evaluated using synthetic and real data.
|
Title: Decomposition of the NVALUE constraint
|
Abstract: We study decompositions of NVALUE, a global constraint that can be used to model a wide range of problems where values need to be counted. Whilst decomposition typically hinders propagation, we identify one decomposition that maintains a global view as enforcing bound consistency on the decomposition achieves bound consistency on the original global NVALUE constraint. Such decompositions offer the prospect for advanced solving techniques like nogood learning and impact based branching heuristics. They may also help SAT and IP solvers take advantage of the propagation of global constraints.
|
Title: Symmetries of Symmetry Breaking Constraints
|
Abstract: Symmetry is an important feature of many constraint programs. We show that any symmetry acting on a set of symmetry breaking constraints can be used to break symmetry. Different symmetries pick out different solutions in each symmetry class. We use these observations in two methods for eliminating symmetry from a problem. These methods are designed to have many of the advantages of symmetry breaking methods that post static symmetry breaking constraint without some of the disadvantages. In particular, the two methods prune the search space using fast and efficient propagation of posted constraints, whilst reducing the conflict between symmetry breaking and branching heuristics. Experimental results show that the two methods perform well on some standard benchmarks.
|
Title: A possible low-level explanation of "temporal dynamics of brightness induction and White's illusion"
|
Abstract: Based upon physiological observation on time dependent orientation selectivity in the cells of macaque's primary visual cortex together with the psychophysical studies on the tuning of orientation detectors in human vision we suggest that time dependence in brightness perception can be accommodated through the time evolution of cortical contribution to the orientation tuning of the ODoG filter responses. A set of Difference of Gaussians functions has been used to mimic the time dependence of orientation tuning. The tuning of orientation preference and its inversion at a later time have been considered in explaining qualitatively the temporal dynamics of brightness perception observed in "Brief presentations reveal the temporal dynamics of brightness induction and White's illusion" for 58 and 82 ms of stimulus exposure.
|
Title: Analyse en d\'ependances \`a l'aide des grammaires d'interaction
|
Abstract: This article proposes a method to extract dependency structures from phrase-structure level parsing with Interaction Grammars. Interaction Grammars are a formalism which expresses interactions among words using a polarity system. Syntactical composition is led by the saturation of polarities. Interactions take place between constituents, but as grammars are lexicalized, these interactions can be translated at the level of words. Dependency relations are extracted from the parsing process: every dependency is the consequence of a polarity saturation. The dependency relations we obtain can be seen as a refinement of the usual dependency tree. Generally speaking, this work sheds new light on links between phrase structure and dependency parsing.
|
Title: Grouping Synonyms by Definitions
|
Abstract: We present a method for grouping the synonyms of a lemma according to its dictionary senses. The senses are defined by a large machine readable dictionary for French, the TLFi (Tr\'esor de la langue franaise informatis\'e) and the synonyms are given by 5 synonym dictionaries (also for French). To evaluate the proposed method, we manually constructed a gold standard where for each (word, definition) pair and given the set of synonyms defined for that word by the 5 synonym dictionaries, 4 lexicographers specified the set of synonyms they judge adequate. While inter-annotator agreement ranges on that task from 67% to at best 88% depending on the annotator pair and on the synonym dictionary being considered, the automatic procedure we propose scores a precision of 67% and a recall of 71%. The proposed method is compared with related work namely, word sense disambiguation, synonym lexicon acquisition and WordNet construction.
|
Title: Mathematics, Recursion, and Universals in Human Languages
|
Abstract: There are many scientific problems generated by the multiple and conflicting alternative definitions of linguistic recursion and human recursive processing that exist in the literature. The purpose of this article is to make available to the linguistic community the standard mathematical definition of recursion and to apply it to discuss linguistic recursion. As a byproduct, we obtain an insight into certain "soft universals" of human languages, which are related to cognitive constructs necessary to implement mathematical reasoning, i.e. mathematical model theory.
|
Title: Exploiting Unlabeled Data to Enhance Ensemble Diversity
|
Abstract: Ensemble learning aims to improve generalization ability by using multiple base learners. It is well-known that to construct a good ensemble, the base learners should be accurate as well as diverse. In this paper, unlabeled data is exploited to facilitate ensemble learning by helping augment the diversity among the base learners. Specifically, a semi-supervised ensemble method named UDEED is proposed. Unlike existing semi-supervised ensemble methods where error-prone pseudo-labels are estimated for unlabeled data to enlarge the labeled data to improve accuracy, UDEED works by maximizing accuracies of base learners on labeled data while maximizing diversity among them on unlabeled data. Experiments show that UDEED can effectively utilize unlabeled data for ensemble learning and is highly competitive to well-established semi-supervised ensemble methods.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.