Convex Optimization Algorithms and Statistical Bounds for Learning Structured Models

Thesis (Ph.D.)--University of Washington, 2016-08

Positive solutions for parametric Dirichlet problems with indefinite potential and superdiffusive reaction

We consider a parametric semilinear Dirichlet problem driven by the Laplacian plus an indefinite unbounded potential and with a reaction of superdifissive type. Using variational and truncation techniques, we show that there exists a critical parameter value λ_{∗}>0 such that for all λ> λ_{∗} the problem has least two positive solutions, for λ= λ_{∗} the problem has at least one positive solutions, and no positive solutions exist when λ∈(0,λ_{∗}). Also, we show that for λ≥ λ_{∗} the problem has a smallest positive solution.

Discrete-event Simulation and Optimization to Improve the Performance of a Healthcare System

Thesis (Ph.D.)--University of Washington, 2016-08

Nonlinear preservers

Dans cette thèse, nous sommes intéressés par des problèmes de préservation des applications non-linéaires entre deux algèbres de Banach complexes unitaires A et B. En général, ces problèmes demandent la caractérisation des applications φ : A → B non nécessairement linéaires, qui laissent invariant une propriété, une relation ou un sous-ensemble. Dans le Chapitre 3, la description des applications surjectives φ de B(X) sur B(Y), qui satisfont c(φ(S)±φ(T)) = c(S ± T), (S,T ∈ B(X)), est donnée, où c(·) représente soit le module minimal, ou le module de surjectivité ou le module maximal et B(X) (resp. B(Y)) dénote l’algèbre de tous les opérateurs linéaires et bornés sur X (resp. sur Y). Dans le Chapitre 4, une question similaire pour la conorme des opérateurs, est considérée. La caractérisation des applications bicontinues et bijectives φ deB(X) surB(Y), qui satisfont γ(φ(S ± φ(T)) = γ(S ± T), (S,T ∈ B(X)), est obtenue. Le Chapitre 5 est consacré à la description des applications surjectives φ1,φ2 d’une algèbre de Banach semisimple A sur une algèbre de Banach B avec un socle essentiel, qui satisfont σ(φ1(a)φ2(b)) = σ(ab), (a,b ∈ A). Aussi, la caractérisation des applications φ de A sur B, sous les mêmes hypothèses sur A et B, qui satisfont σ(φ(a)φ(b)φ(a)) = σ(aba), (a,b ∈ A), est donnée. Comme conséquences, nous incluons les résultats obtenus au cas des algèbres B(X) et B(Y).

Problems in Algebraic Vision

Thesis (Ph.D.)--University of Washington, 2016-08

Ant Colony Optimization for Requirement selection in Incremental Software development

This proposal shows that ACO systems can be applied to problems of requirements selection in software incremental development, with the idea of obtaining better results of those produced by expert judgment alone. The evaluation of the ACO systems should be done through a compared analysis with greedy and simulated annealing algorithms, performing experiments with some problems instances

'The application of Bayesian Optimization and Classifier Systems in Nurse Scheduling'

Abstract. Two ideas taken from Bayesian optimization and classifier systems are presented for personnel scheduling based on choosing a suitable scheduling rule from a set for each person's assignment. Unlike our previous work of using genetic algorithms whose learning is implicit, the learning in both approaches is explicit, i.e. we are able to identify building blocks directly. To achieve this target, the Bayesian optimization algorithm builds a Bayesian network of the joint probability distribution of the rules used to construct solutions, while the adapted classifier system assigns each rule a strength value that is constantly updated according to its usefulness in the current situation. Computational results from 52 real data instances of nurse scheduling demonstrate the success of both approaches. It is also suggested that the learning mechanism in the proposed approaches might be suitable for other scheduling problems.

Nonlinear distortion generation mechanisms in Doherty amplifiers

Nowadays, wireless communications systems demand for greater mobility and higher data rates. Moreover, the need for spectral efficiency requires the use of non-constant envelope modulation schemes. Hence, power amplifier designers have to build highly efficient, broadband and linear amplifiers. In order to fulfil these strict requirements, the practical Doherty amplifier seems to be the most promising technique. However, due to its complex operation, its nonlinear distortion generation mechanisms are not yet fully understood. Currently, only heuristic interpretations are being used to justify the observed phenomena. Therefore, the main objective of this work is to provide a model capable of describing the Doherty power amplifier nonlinear distortion generation mechanisms, allowing the optimization of its design according to linearity and efficiency criteria. Besides that, this approach will allow a bridge between two different worlds: power amplifier design and digital pre-distortion since the knowledge gathered from the Doherty operation will serve to select the most suitable pre-distortion models.

Hybrid Graph Heuristics within a Hyper-heuristic Approach to Exam Timetabling Problems

This paper is concerned with the hybridization of two graph coloring heuristics (Saturation Degree and Largest Degree), and their application within a hyperheuristic for exam timetabling problems. Hyper-heuristics can be seen as algorithms which intelligently select appropriate algorithms/heuristics for solving a problem. We developed a Tabu Search based hyper-heuristic to search for heuristic lists (of graph heuristics) for solving problems and investigated the heuristic lists found by employing knowledge discovery techniques. Two hybrid approaches (involving Saturation Degree and Largest Degree) including one which employs Case Based Reasoning are presented and discussed. Both the Tabu Search based hyper-heuristic and the hybrid approaches are tested on random and real-world exam timetabling problems. Experimental results are comparable with the best state-of-the-art approaches (as measured against established benchmark problems). The results also demonstrate an increased level of generality in our approach.

Digital nonlinear equalization for optical transmission systems

This thesis focuses on digital equalization of nonlinear fiber impairments for coherent optical transmission systems. Building from well-known physical models of signal propagation in single-mode optical fibers, novel nonlinear equalization techniques are proposed, numerically assessed and experimentally demonstrated. The structure of the proposed algorithms is strongly driven by the optimization of the performance versus complexity tradeoff, envisioning the near-future practical application in commercial real-time transceivers. The work is initially focused on the mitigation of intra-channel nonlinear impairments relying on the concept of digital backpropagation (DBP) associated with Volterra-based filtering. After a comprehensive analysis of the third-order Volterra kernel, a set of critical simplifications are identified, culminating in the development of reduced complexity nonlinear equalization algorithms formulated both in time and frequency domains. The implementation complexity of the proposed techniques is analytically described in terms of computational effort and processing latency, by determining the number of real multiplications per processed sample and the number of serial multiplications, respectively. The equalization performance is numerically and experimentally assessed through bit error rate (BER) measurements. Finally, the problem of inter-channel nonlinear compensation is addressed within the context of 400 Gb/s (400G) superchannels for long-haul and ultra-long-haul transmission. Different superchannel configurations and nonlinear equalization strategies are experimentally assessed, demonstrating that inter-subcarrier nonlinear equalization can provide an enhanced signal reach while requiring only marginal added complexity.

A Multi-Objective Approach for VNE: Problems using multiple ILP formulations

Network Virtualization is a key technology for the Future Internet, allowing the deployment of multiple independent virtual networks that use resources of the same basic infrastructure. An important challenge in the dynamic provision of virtual networks resides in the optimal allocation of physical resources (nodes and links) to requirements of virtual networks. This problem is known as Virtual Network Embedding (VNE). For the resolution of this problem, previous research has focused on designing algorithms based on the optimization of a single objective. On the contrary, in this work we present a multi-objective algorithm, called VNE-MO-ILP, for solving dynamic VNE problem, which calculates an approximation of the Pareto Front considering simultaneously resource utilization and load balancing. Experimental results show evidences that the proposed algorithm is better or at least comparable to a state-of-the-art algorithm. Two performance metrics were simultaneously evaluated: (i) Virtual Network Request Acceptance Ratio and (ii) Revenue/Cost Relation. The size of test networks used in the experiments shows that the proposed algorithm scales well in execution times, for networks of 84 nodes

Variational ensemble Kalman filtering applied to data assimilation problems in computational fluid dynamics

One challenge on data assimilation (DA) methods is how the error covariance for the model state is computed. Ensemble methods have been proposed for producing error covariance estimates, as error is propagated in time using the non-linear model. Variational methods, on the other hand, use the concepts of control theory, whereby the state estimate is optimized from both the background and the measurements. Numerical optimization schemes are applied which solve the problem of memory storage and huge matrix inversion needed by classical Kalman filter methods. Variational Ensemble Kalman filter (VEnKF), as a method inspired the Variational Kalman Filter (VKF), enjoys the benefits from both ensemble methods and variational methods. It avoids filter inbreeding problems which emerge when the ensemble spread underestimates the true error covariance. In VEnKF this is tackled by resampling the ensemble every time measurements are available. One advantage of VEnKF over VKF is that it needs neither tangent linear code nor adjoint code. In this thesis, VEnKF has been applied to a two-dimensional shallow water model simulating a dam-break experiment. The model is a public code with water height measurements recorded in seven stations along the 21:2 m long 1:4 m wide flume’s mid-line. Because the data were too sparse to assimilate the 30 171 model state vector, we chose to interpolate the data both in time and in space. The results of the assimilation were compared with that of a pure simulation. We have found that the results revealed by the VEnKF were more realistic, without numerical artifacts present in the pure simulation. Creating a wrapper code for a model and DA scheme might be challenging, especially when the two were designed independently or are poorly documented. In this thesis we have presented a non-intrusive approach of coupling the model and a DA scheme. An external program is used to send and receive information between the model and DA procedure using files. The advantage of this method is that the model code changes needed are minimal, only a few lines which facilitate input and output. Apart from being simple to coupling, the approach can be employed even if the two were written in different programming languages, because the communication is not through code. The non-intrusive approach is made to accommodate parallel computing by just telling the control program to wait until all the processes have ended before the DA procedure is invoked. It is worth mentioning the overhead increase caused by the approach, as at every assimilation cycle both the model and the DA procedure have to be initialized. Nonetheless, the method can be an ideal approach for a benchmark platform in testing DA methods. The non-intrusive VEnKF has been applied to a multi-purpose hydrodynamic model COHERENS to assimilate Total Suspended Matter (TSM) in lake Säkylän Pyhäjärvi. The lake has an area of 154 km2 with an average depth of 5:4 m. Turbidity and chlorophyll-a concentrations from MERIS satellite images for 7 days between May 16 and July 6 2009 were available. The effect of the organic matter has been computationally eliminated to obtain TSM data. Because of computational demands from both COHERENS and VEnKF, we have chosen to use 1 km grid resolution. The results of the VEnKF have been compared with the measurements recorded at an automatic station located at the North-Western part of the lake. However, due to TSM data sparsity in both time and space, it could not be well matched. The use of multiple automatic stations with real time data is important to elude the time sparsity problem. With DA, this will help in better understanding the environmental hazard variables for instance. We have found that using a very high ensemble size does not necessarily improve the results, because there is a limit whereby additional ensemble members add very little to the performance. Successful implementation of the non-intrusive VEnKF and the ensemble size limit for performance leads to an emerging area of Reduced Order Modeling (ROM). To save computational resources, running full-blown model in ROM is avoided. When the ROM is applied with the non-intrusive DA approach, it might result in a cheaper algorithm that will relax computation challenges existing in the field of modelling and DA.

Bayesian Optimisation Algorithm for Nurse Scheduling, Scalable Optimization via Probabilistic Modeling

Our research has shown that schedules can be built mimicking a human scheduler by using a set of rules that involve domain knowledge. This chapter presents a Bayesian Optimization Algorithm (BOA) for the nurse scheduling problem that chooses such suitable scheduling rules from a set for each nurse’s assignment. Based on the idea of using probabilistic models, the BOA builds a Bayesian network for the set of promising solutions and samples these networks to generate new candidate solutions. Computational results from 52 real data instances demonstrate the success of this approach. It is also suggested that the learning mechanism in the proposed algorithm may be suitable for other scheduling problems.

'The application of Bayesian Optimization and Classifier Systems in Nurse Scheduling'

Abstract. Two ideas taken from Bayesian optimization and classifier systems are presented for personnel scheduling based on choosing a suitable scheduling rule from a set for each person's assignment. Unlike our previous work of using genetic algorithms whose learning is implicit, the learning in both approaches is explicit, i.e. we are able to identify building blocks directly. To achieve this target, the Bayesian optimization algorithm builds a Bayesian network of the joint probability distribution of the rules used to construct solutions, while the adapted classifier system assigns each rule a strength value that is constantly updated according to its usefulness in the current situation. Computational results from 52 real data instances of nurse scheduling demonstrate the success of both approaches. It is also suggested that the learning mechanism in the proposed approaches might be suitable for other scheduling problems.

Dynamic Programming Approaches for Estimating and Applying Large-scale Discrete Choice Models

People go through their life making all kinds of decisions, and some of these decisions affect their demand for transportation, for example, their choices of where to live and where to work, how and when to travel and which route to take. Transport related choices are typically time dependent and characterized by large number of alternatives that can be spatially correlated. This thesis deals with models that can be used to analyze and predict discrete choices in large-scale networks. The proposed models and methods are highly relevant for, but not limited to, transport applications. We model decisions as sequences of choices within the dynamic discrete choice framework, also known as parametric Markov decision processes. Such models are known to be difficult to estimate and to apply to make predictions because dynamic programming problems need to be solved in order to compute choice probabilities. In this thesis we show that it is possible to explore the network structure and the flexibility of dynamic programming so that the dynamic discrete choice modeling approach is not only useful to model time dependent choices, but also makes it easier to model large-scale static choices. The thesis consists of seven articles containing a number of models and methods for estimating, applying and testing large-scale discrete choice models. In the following we group the contributions under three themes: route choice modeling, large-scale multivariate extreme value (MEV) model estimation and nonlinear optimization algorithms. Five articles are related to route choice modeling. We propose different dynamic discrete choice models that allow paths to be correlated based on the MEV and mixed logit models. The resulting route choice models become expensive to estimate and we deal with this challenge by proposing innovative methods that allow to reduce the estimation cost. For example, we propose a decomposition method that not only opens up for possibility of mixing, but also speeds up the estimation for simple logit models, which has implications also for traffic simulation. Moreover, we compare the utility maximization and regret minimization decision rules, and we propose a misspecification test for logit-based route choice models. The second theme is related to the estimation of static discrete choice models with large choice sets. We establish that a class of MEV models can be reformulated as dynamic discrete choice models on the networks of correlation structures. These dynamic models can then be estimated quickly using dynamic programming techniques and an efficient nonlinear optimization algorithm. Finally, the third theme focuses on structured quasi-Newton techniques for estimating discrete choice models by maximum likelihood. We examine and adapt switching methods that can be easily integrated into usual optimization algorithms (line search and trust region) to accelerate the estimation process. The proposed dynamic discrete choice models and estimation methods can be used in various discrete choice applications. In the area of big data analytics, models that can deal with large choice sets and sequential choices are important. Our research can therefore be of interest in various demand analysis applications (predictive analytics) or can be integrated with optimization models (prescriptive analytics). Furthermore, our studies indicate the potential of dynamic programming techniques in this context, even for static models, which opens up a variety of future research directions.