Advances in scaling deep learning algorithms

Les algorithmes d'apprentissage profond forment un nouvel ensemble de méthodes puissantes pour l'apprentissage automatique. L'idée est de combiner des couches de facteurs latents en hierarchies. Cela requiert souvent un coût computationel plus elevé et augmente aussi le nombre de paramètres du modèle. Ainsi, l'utilisation de ces méthodes sur des problèmes à plus grande échelle demande de réduire leur coût et aussi d'améliorer leur régularisation et leur optimization. Cette thèse adresse cette question sur ces trois perspectives. Nous étudions tout d'abord le problème de réduire le coût de certains algorithmes profonds. Nous proposons deux méthodes pour entrainer des machines de Boltzmann restreintes et des auto-encodeurs débruitants sur des distributions sparses à haute dimension. Ceci est important pour l'application de ces algorithmes pour le traitement de langues naturelles. Ces deux méthodes (Dauphin et al., 2011; Dauphin and Bengio, 2013) utilisent l'échantillonage par importance pour échantilloner l'objectif de ces modèles. Nous observons que cela réduit significativement le temps d'entrainement. L'accéleration atteint 2 ordres de magnitude sur plusieurs bancs d'essai. Deuxièmement, nous introduisont un puissant régularisateur pour les méthodes profondes. Les résultats expérimentaux démontrent qu'un bon régularisateur est crucial pour obtenir de bonnes performances avec des gros réseaux (Hinton et al., 2012). Dans Rifai et al. (2011), nous proposons un nouveau régularisateur qui combine l'apprentissage non-supervisé et la propagation de tangente (Simard et al., 1992). Cette méthode exploite des principes géometriques et permit au moment de la publication d'atteindre des résultats à l'état de l'art. Finalement, nous considérons le problème d'optimiser des surfaces non-convexes à haute dimensionalité comme celle des réseaux de neurones. Tradionellement, l'abondance de minimum locaux était considéré comme la principale difficulté dans ces problèmes. Dans Dauphin et al. (2014a) nous argumentons à partir de résultats en statistique physique, de la théorie des matrices aléatoires, de la théorie des réseaux de neurones et à partir de résultats expérimentaux qu'une difficulté plus profonde provient de la prolifération de points-selle. Dans ce papier nous proposons aussi une nouvelle méthode pour l'optimisation non-convexe.

Transformations quasi-conformes de maillages volumiques et applications en infographie

La modélisation géométrique est importante autant en infographie qu'en ingénierie. Notre capacité à représenter l'information géométrique fixe les limites et la facilité avec laquelle on manipule les objets 3D. Une de ces représentations géométriques est le maillage volumique, formé de polyèdres assemblés de sorte à approcher une forme désirée. Certaines applications, tels que le placage de textures et le remaillage, ont avantage à déformer le maillage vers un domaine plus régulier pour faciliter le traitement. On dit qu'une déformation est \emph{quasi-conforme} si elle borne la distorsion. Cette thèse porte sur l’étude et le développement d'algorithmes de déformation quasi-conforme de maillages volumiques. Nous étudions ces types de déformations parce qu’elles offrent de bonnes propriétés de préservation de l’aspect local d’un solide et qu’elles ont été peu étudiées dans le contexte de l’informatique graphique, contrairement à leurs pendants 2D. Cette recherche tente de généraliser aux volumes des concepts bien maitrisés pour la déformation de surfaces. Premièrement, nous présentons une approche linéaire de la quasi-conformité. Nous développons une méthode déformant l’objet vers son domaine paramétrique par une méthode des moindres carrés linéaires. Cette méthode est simple d'implémentation et rapide d'exécution, mais n'est qu'une approximation de la quasi-conformité car elle ne borne pas la distorsion. Deuxièmement, nous remédions à ce problème par une approche non linéaire basée sur les positions des sommets. Nous développons une technique déformant le domaine paramétrique vers le solide par une méthode des moindres carrés non linéaires. La non-linéarité permet l’inclusion de contraintes garantissant l’injectivité de la déformation. De plus, la déformation du domaine paramétrique au lieu de l’objet lui-même permet l’utilisation de domaines plus généraux. Troisièmement, nous présentons une approche non linéaire basée sur les angles dièdres. Cette méthode définit la déformation du solide par les angles dièdres au lieu des positions des sommets du maillage. Ce changement de variables permet une expression naturelle des bornes de distorsion de la déformation. Nous présentons quelques applications de cette nouvelle approche dont la paramétrisation, l'interpolation, l'optimisation et la compression de maillages tétraédriques.

Analysis of Some Stochastic Inventory Models with Pooling Retrial of Customers

The thesis deals with analysis of some Stochastic Inventory Models with Pooling/Retrial of Customers.. In the first model we analyze an (s,S) production Inventory system with retrial of customers. Arrival of customers from outside the system form a Poisson process. The inter production times are exponentially distributed with parameter µ. When inventory level reaches zero further arriving demands are sent to the orbit which has capacity M(<∞). Customers, who find the orbit full and inventory level at zero are lost to the system. Demands arising from the orbital customers are exponentially distributed with parameter γ. In the model-II we extend these results to perishable inventory system assuming that the life-time of each item follows exponential with parameter θ. The study deals with an (s,S) production inventory with service times and retrial of unsatisfied customers. Primary demands occur according to a Markovian Arrival Process(MAP). Consider an (s,S)-retrial inventory with service time in which primary demands occur according to a Batch Markovian Arrival Process (BMAP). The inventory is controlled by the (s,S) policy and (s,S) inventory system with service time. Primary demands occur according to Poissson process with parameter λ. The study concentrates two models. In the first model we analyze an (s,S) Inventory system with postponed demands where arrivals of demands form a Poisson process. In the second model, we extend our results to perishable inventory system assuming that the life-time of each item follows exponential distribution with parameter θ. Also it is assumed that when inventory level is zero the arriving demands choose to enter the pool with probability β and with complementary probability (1- β) it is lost for ever. Finally it analyze an (s,S) production inventory system with switching time. A lot of work is reported under the assumption that the switching time is negligible but this is not the case for several real life situation.

ON APPROXIMATION METHODS FOR UNBOUNDED OPERATORS

Department of Mathematics, Cochin University of Science and Technology

Use of Novel Algorithms MAJE4 and MACJER-320 for Achieving Confidentiality and Message Authentication in SSL & TLS

Extensive use of the Internet coupled with the marvelous growth in e-commerce and m-commerce has created a huge demand for information security. The Secure Socket Layer (SSL) protocol is the most widely used security protocol in the Internet which meets this demand. It provides protection against eaves droppings, tampering and forgery. The cryptographic algorithms RC4 and HMAC have been in use for achieving security services like confidentiality and authentication in the SSL. But recent attacks against RC4 and HMAC have raised questions in the confidence on these algorithms. Hence two novel cryptographic algorithms MAJE4 and MACJER-320 have been proposed as substitutes for them. The focus of this work is to demonstrate the performance of these new algorithms and suggest them as dependable alternatives to satisfy the need of security services in SSL. The performance evaluation has been done by using practical implementation method.

Gamma stochastic volatility models

This paper presents gamma stochastic volatility models and investigates its distributional and time series properties. The parameter estimators obtained by the method of moments are shown analytically to be consistent and asymptotically normal. The simulation results indicate that the estimators behave well. The insample analysis shows that return models with gamma autoregressive stochastic volatility processes capture the leptokurtic nature of return distributions and the slowly decaying autocorrelation functions of squared stock index returns for the USA and UK. In comparison with GARCH and EGARCH models, the gamma autoregressive model picks up the persistence in volatility for the US and UK index returns but not the volatility persistence for the Canadian and Japanese index returns. The out-of-sample analysis indicates that the gamma autoregressive model has a superior volatility forecasting performance compared to GARCH and EGARCH models.

Stochastic Modelling Using Markov Chains

Department of Statistics, Cochin University of Science and Technology

An approximation method for evaluating centrifugal distortion constants in XH2 bent symmetric molecules

University of Cochin

Studies,Design and Development of Network Security Enhancement Services Using Novel Cryptographic Algorithms

Internet today has become a vital part of day to day life, owing to the revolutionary changes it has brought about in various fields. Dependence on the Internet as an information highway and knowledge bank is exponentially increasing so that a going back is beyond imagination. Transfer of critical information is also being carried out through the Internet. This widespread use of the Internet coupled with the tremendous growth in e-commerce and m-commerce has created a vital need for infonnation security.Internet has also become an active field of crackers and intruders. The whole development in this area can become null and void if fool-proof security of the data is not ensured without a chance of being adulterated. It is, hence a challenge before the professional community to develop systems to ensure security of the data sent through the Internet.Stream ciphers, hash functions and message authentication codes play vital roles in providing security services like confidentiality, integrity and authentication of the data sent through the Internet. There are several ·such popular and dependable techniques, which have been in use widely, for quite a long time. This long term exposure makes them vulnerable to successful or near successful attempts for attacks. Hence it is the need of the hour to develop new algorithms with better security.Hence studies were conducted on various types of algorithms being used in this area. Focus was given to identify the properties imparting security at this stage. By making use of a perception derived from these studies, new algorithms were designed. Performances of these algorithms were then studied followed by necessary modifications to yield an improved system consisting of a new stream cipher algorithm MAJE4, a new hash code JERIM- 320 and a new message authentication code MACJER-320. Detailed analysis and comparison with the existing popular schemes were also carried out to establish the security levels.The Secure Socket Layer (SSL) I Transport Layer Security (TLS) protocol is one of the most widely used security protocols in Internet. The cryptographic algorithms RC4 and HMAC have been in use for achieving security services like confidentiality and authentication in the SSL I TLS. But recent attacks on RC4 and HMAC have raised questions about the reliability of these algorithms. Hence MAJE4 and MACJER-320 have been proposed as substitutes for them. Detailed studies on the performance of these new algorithms were carried out; it has been observed that they are dependable alternatives.

Mean Squared Residue Based Biclustering Algorithms for the Analysis of Gene Expression Data

Computational Biology is the research are that contributes to the analysis of biological data through the development of algorithms which will address significant research problems.The data from molecular biology includes DNA,RNA ,Protein and Gene expression data.Gene Expression Data provides the expression level of genes under different conditions.Gene expression is the process of transcribing the DNA sequence of a gene into mRNA sequences which in turn are later translated into proteins.The number of copies of mRNA produced is called the expression level of a gene.Gene expression data is organized in the form of a matrix. Rows in the matrix represent genes and columns in the matrix represent experimental conditions.Experimental conditions can be different tissue types or time points.Entries in the gene expression matrix are real values.Through the analysis of gene expression data it is possible to determine the behavioral patterns of genes such as similarity of their behavior,nature of their interaction,their respective contribution to the same pathways and so on. Similar expression patterns are exhibited by the genes participating in the same biological process.These patterns have immense relevance and application in bioinformatics and clinical research.Theses patterns are used in the medical domain for aid in more accurate diagnosis,prognosis,treatment planning.drug discovery and protein network analysis.To identify various patterns from gene expression data,data mining techniques are essential.Clustering is an important data mining technique for the analysis of gene expression data.To overcome the problems associated with clustering,biclustering is introduced.Biclustering refers to simultaneous clustering of both rows and columns of a data matrix. Clustering is a global whereas biclustering is a local model.Discovering local expression patterns is essential for identfying many genetic pathways that are not apparent otherwise.It is therefore necessary to move beyond the clustering paradigm towards developing approaches which are capable of discovering local patterns in gene expression data.A biclusters is a submatrix of the gene expression data matrix.The rows and columns in the submatrix need not be contiguous as in the gene expression data matrix.Biclusters are not disjoint.Computation of biclusters is costly because one will have to consider all the combinations of columans and rows in order to find out all the biclusters.The search space for the biclustering problem is 2 m+n where m and n are the number of genes and conditions respectively.Usually m+n is more than 3000.The biclustering problem is NP-hard.Biclustering is a powerful analytical tool for the biologist.The research reported in this thesis addresses the problem of biclustering.Ten algorithms are developed for the identification of coherent biclusters from gene expression data.All these algorithms are making use of a measure called mean squared residue to search for biclusters.The objective here is to identify the biclusters of maximum size with the mean squared residue lower than a given threshold. All these algorithms begin the search from tightly coregulated submatrices called the seeds.These seeds are generated by K-Means clustering algorithm.The algorithms developed can be classified as constraint based,greedy and metaheuristic.Constarint based algorithms uses one or more of the various constaints namely the MSR threshold and the MSR difference threshold.The greedy approach makes a locally optimal choice at each stage with the objective of finding the global optimum.In metaheuristic approaches particle Swarm Optimization(PSO) and variants of Greedy Randomized Adaptive Search Procedure(GRASP) are used for the identification of biclusters.These algorithms are implemented on the Yeast and Lymphoma datasets.Biologically relevant and statistically significant biclusters are identified by all these algorithms which are validated by Gene Ontology database.All these algorithms are compared with some other biclustering algorithms.Algorithms developed in this work overcome some of the problems associated with the already existing algorithms.With the help of some of the algorithms which are developed in this work biclusters with very high row variance,which is higher than the row variance of any other algorithm using mean squared residue, are identified from both Yeast and Lymphoma data sets.Such biclusters which make significant change in the expression level are highly relevant biologically.

Spin and density longitudinal response of quantum dots in the time-dependent local-spin-density approximation

The longitudinal dipole response of a quantum dot has been calculated in the far-infrared regime using local-spin-density-functional theory. We have studied the coupling between the collective spin and density modes as a function of the magnetic field. We have found that the spin dipole mode and single-particle excitations have a sizable overlap, and that the magnetoplasmon modes can be excited by the dipole spin operator if the dot is spin polarized. The frequency of the dipole spin edge mode presents an oscillation which is clearly filling factor (v) related. We have found that the spin dipole mode is especially soft for even-n values. Results for selected numbers of electrons and confining potentials are discussed.

Stochastic Inventory with/without Service Time, Reneging and Retrial of Customers

This thesis Entitled Stochastic modelling and analysis.This thesis is divided into six chapters including this introductory chapter. In second chapter, we consider an (s,S) inventory model with service, reneging of customers and finite shortage of items.In the third chapter, we consider an (s,S) inventoiy system with retrial of customers. Arrival of customers forms a Poisson process with rate. When the inventory level depletes to s due to demands, an order for replenishment is placed.In Chapter 4, we analyze and compare three (s,S) inventory systems with positive service time and retrial of customers. In all these systems, arrivals of customers form a Poisson process and service times are exponentially distributed. In chapter 5, we analyze and compare three production inventory systems with positive service time and retrial of customers. In all these systems, arrivals of customers form a Poisson process and service times are exponentially distributed.In chapter 6, we consider a PH /PH /l inventory model with reneging of customers and finite shortage of items.