Learning to Interpret and Generate Instructional Recipes

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

Convex Optimization Algorithms and Statistical Bounds for Learning Structured Models

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

IntelliChair: a non-intrusive sitting posture and sitting activity recognition system

Current Ambient Intelligence and Intelligent Environment research focuses on the interpretation of a subject’s behaviour at the activity level by logging the Activity of Daily Living (ADL) such as eating, cooking, etc. In general, the sensors employed (e.g. PIR sensors, contact sensors) provide low resolution information. Meanwhile, the expansion of ubiquitous computing allows researchers to gather additional information from different types of sensor which is possible to improve activity analysis. Based on the previous research about sitting posture detection, this research attempts to further analyses human sitting activity. The aim of this research is to use non-intrusive low cost pressure sensor embedded chair system to recognize a subject’s activity by using their detected postures. There are three steps for this research, the first step is to find a hardware solution for low cost sitting posture detection, second step is to find a suitable strategy of sitting posture detection and the last step is to correlate the time-ordered sitting posture sequences with sitting activity. The author initiated a prototype type of sensing system called IntelliChair for sitting posture detection. Two experiments are proceeded in order to determine the hardware architecture of IntelliChair system. The prototype looks at the sensor selection and integration of various sensor and indicates the best for a low cost, non-intrusive system. Subsequently, this research implements signal process theory to explore the frequency feature of sitting posture, for the purpose of determining a suitable sampling rate for IntelliChair system. For second and third step, ten subjects are recruited for the sitting posture data and sitting activity data collection. The former dataset is collected byasking subjects to perform certain pre-defined sitting postures on IntelliChair and it is used for posture recognition experiment. The latter dataset is collected by asking the subjects to perform their normal sitting activity routine on IntelliChair for four hours, and the dataset is used for activity modelling and recognition experiment. For the posture recognition experiment, two Support Vector Machine (SVM) based classifiers are trained (one for spine postures and the other one for leg postures), and their performance evaluated. Hidden Markov Model is utilized for sitting activity modelling and recognition in order to establish the selected sitting activities from sitting posture sequences.2. After experimenting with possible sensors, Force Sensing Resistor (FSR) is selected as the pressure sensing unit for IntelliChair. Eight FSRs are mounted on the seat and back of a chair to gather haptic (i.e., touch-based) posture information. Furthermore, the research explores the possibility of using alternative non-intrusive sensing technology (i.e. vision based Kinect Sensor from Microsoft) and find out the Kinect sensor is not reliable for sitting posture detection due to the joint drifting problem. A suitable sampling rate for IntelliChair is determined according to the experiment result which is 6 Hz. The posture classification performance shows that the SVM based classifier is robust to “familiar” subject data (accuracy is 99.8% with spine postures and 99.9% with leg postures). When dealing with “unfamiliar” subject data, the accuracy is 80.7% for spine posture classification and 42.3% for leg posture classification. The result of activity recognition achieves 41.27% accuracy among four selected activities (i.e. relax, play game, working with PC and watching video). The result of this thesis shows that different individual body characteristics and sitting habits influence both sitting posture and sitting activity recognition. In this case, it suggests that IntelliChair is suitable for individual usage but a training stage is required.

Techniques for batch reinforcement learning in robotics

This thesis addresses the Batch Reinforcement Learning methods in Robotics. This sub-class of Reinforcement Learning has shown promising results and has been the focus of recent research. Three contributions are proposed that aim to extend the state-of-art methods allowing for a faster and more stable learning process, such as required for learning in Robotics. The Q-learning update-rule is widely applied, since it allows to learn without the presence of a model of the environment. However, this update-rule is transition-based and does not take advantage of the underlying episodic structure of collected batch of interactions. The Q-Batch update-rule is proposed in this thesis, to process experiencies along the trajectories collected in the interaction phase. This allows a faster propagation of obtained rewards and penalties, resulting in faster and more robust learning. Non-parametric function approximations are explored, such as Gaussian Processes. This type of approximators allows to encode prior knowledge about the latent function, in the form of kernels, providing a higher level of exibility and accuracy. The application of Gaussian Processes in Batch Reinforcement Learning presented a higher performance in learning tasks than other function approximations used in the literature. Lastly, in order to extract more information from the experiences collected by the agent, model-learning techniques are incorporated to learn the system dynamics. In this way, it is possible to augment the set of collected experiences with experiences generated through planning using the learned models. Experiments were carried out mainly in simulation, with some tests carried out in a physical robotic platform. The obtained results show that the proposed approaches are able to outperform the classical Fitted Q Iteration.

A Research of Resource Scheduling Algorithm Based on Hadoop

With the development of electronic devices, more and more mobile clients are connected to the Internet and they generate massive data every day. We live in an age of “Big Data”, and every day we generate hundreds of million magnitude data. By analyzing the data and making prediction, we can carry out better development plan. Unfortunately, traditional computation framework cannot meet the demand, so the Hadoop would be put forward. First the paper introduces the background and development status of Hadoop, compares the MapReduce in Hadoop 1.0 and YARN in Hadoop 2.0, and analyzes the advantages and disadvantages of them. Because the resource management module is the core role of YARN, so next the paper would research about the resource allocation module including the resource management, resource allocation algorithm, resource preemption model and the whole resource scheduling process from applying resource to finishing allocation. Also it would introduce the FIFO Scheduler, Capacity Scheduler, and Fair Scheduler and compare them. The main work has been done in this paper is researching and analyzing the Dominant Resource Fair algorithm of YARN, putting forward a maximum resource utilization algorithm based on Dominant Resource Fair algorithm. The paper also provides a suggestion to improve the unreasonable facts in resource preemption model. Emphasizing “fairness” during resource allocation is the core concept of Dominant Resource Fair algorithm of YARM. Because the cluster is multiple users and multiple resources, so the user’s resource request is multiple too. The DRF algorithm would divide the user’s resources into dominant resource and normal resource. For a user, the dominant resource is the one whose share is highest among all the request resources, others are normal resource. The DRF algorithm requires the dominant resource share of each user being equal. But for these cases where different users’ dominant resource amount differs greatly, emphasizing “fairness” is not suitable and can’t promote the resource utilization of the cluster. By analyzing these cases, this thesis puts forward a new allocation algorithm based on DRF. The new algorithm takes the “fairness” into consideration but not the main principle. Maximizing the resource utilization is the main principle and goal of the new algorithm. According to comparing the result of the DRF and new algorithm based on DRF, we found that the new algorithm has more high resource utilization than DRF. The last part of the thesis is to install the environment of YARN and use the Scheduler Load Simulator (SLS) to simulate the cluster environment.

Requirement Risk Level Forecast Using Bayesian Networks Classifiers

Requirement engineering is a key issue in the development of a software project. Like any other development activity it is not without risks. This work is about the empirical study of risks of requirements by applying machine learning techniques, specifically Bayesian networks classifiers. We have defined several models to predict the risk level for a given requirement using three dataset that collect metrics taken from the requirement specifications of different projects. The classification accuracy of the Bayesian models obtained is evaluated and compared using several classification performance measures. The results of the experiments show that the Bayesians networks allow obtaining valid predictors. Specifically, a tree augmented network structure shows a competitive experimental performance in all datasets. Besides, the relations established between the variables collected to determine the level of risk in a requirement, match with those set by requirement engineers. We show that Bayesian networks are valid tools for the automation of risks assessment in requirement engineering.

Computerized Medical Diagnosis of Melanocytic Lesions based on the ABCD approach

Melanoma is a type of skin cancer and is caused by the uncontrolled growth of atypical melanocytes. In recent decades, computer aided diagnosis is used to support medical professionals; however, there is still no globally accepted tool. In this context, similar to state-of-the-art we propose a system that receives a dermatoscopy image and provides a diagnostic if the lesion is benign or malignant. This tool is composed with next modules: Preprocessing, Segmentation, Feature Extraction, and Classification. Preprocessing involves the removal of hairs. Segmentation is to isolate the lesion. Feature extraction is considering the ABCD dermoscopy rule. The classification is performed by the Support Vector Machine. Experimental evidence indicates that the proposal has 90.63 % accuracy, 95 % sensitivity, and 83.33 % specificity on a data-set of 104 dermatoscopy images. These results are favorable considering the performance of diagnosis by traditional progress in the area of dermatology

Identifying player profiles in video games

Tämä diplomityö tarkastelee pelaajatyyppien ja pelaajamotivaatioiden tunnistamista videopeleissä. Aiempi tutkimus tuntee monia pelaajatyyppien malleja, mutta niitä ei ole liiemmin sovellettu käytäntöön peleissä. Tässä työssä suoritetaan systemaattinen kirjallisuuskartoitus erilaisista pelaajatyyppien malleista, jonka pohjalta esitetään useita pelaajien luokittelutapoja. Lisäksi toteutetaan tapaustutkimus, jossa kirjallisuuden pohjalta valitaan pelaajien luokittelumalli ja testataan mallia käytännössä tunnistamalla pelaajatyyppejä data-analytiikan avulla reaaliaikaisessa strategiapelissä.

Spectral Frame Analysis and Learning through Graph Structure

This dissertation investigates the connection between spectral analysis and frame theory. When considering the spectral properties of a frame, we present a few novel results relating to the spectral decomposition. We first show that scalable frames have the property that the inner product of the scaling coefficients and the eigenvectors must equal the inverse eigenvalues. From this, we prove a similar result when an approximate scaling is obtained. We then focus on the optimization problems inherent to the scalable frames by first showing that there is an equivalence between scaling a frame and optimization problems with a non-restrictive objective function. Various objective functions are considered, and an analysis of the solution type is presented. For linear objectives, we can encourage sparse scalings, and with barrier objective functions, we force dense solutions. We further consider frames in high dimensions, and derive various solution techniques. From here, we restrict ourselves to various frame classes, to add more specificity to the results. Using frames generated from distributions allows for the placement of probabilistic bounds on scalability. For discrete distributions (Bernoulli and Rademacher), we bound the probability of encountering an ONB, and for continuous symmetric distributions (Uniform and Gaussian), we show that symmetry is retained in the transformed domain. We also prove several hyperplane-separation results. With the theory developed, we discuss graph applications of the scalability framework. We make a connection with graph conditioning, and show the in-feasibility of the problem in the general case. After a modification, we show that any complete graph can be conditioned. We then present a modification of standard PCA (robust PCA) developed by Cand\`es, and give some background into Electron Energy-Loss Spectroscopy (EELS). We design a novel scheme for the processing of EELS through robust PCA and least-squares regression, and test this scheme on biological samples. Finally, we take the idea of robust PCA and apply the technique of kernel PCA to perform robust manifold learning. We derive the problem and present an algorithm for its solution. There is also discussion of the differences with RPCA that make theoretical guarantees difficult.

Dendritic Cells for Anomaly Detection

Artificial immune systems, more specifically the negative selection algorithm, have previously been applied to intrusion detection. The aim of this research is to develop an intrusion detection system based on a novel concept in immunology, the Danger Theory. Dendritic Cells (DCs) are antigen presenting cells and key to the activation of the human immune system. DCs perform the vital role of combining signals from the host tissue and correlate these signals with proteins known as antigens. In algorithmic terms, individual DCs perform multi-sensor data fusion based on time-windows. The whole population of DCs asynchronously correlates the fused signals with a secondary data stream. The behaviour of human DCs is abstracted to form the DC Algorithm (DCA), which is implemented using an immune inspired framework, libtissue. This system is used to detect context switching for a basic machine learning dataset and to detect outgoing portscans in real-time. Experimental results show a significant difference between an outgoing portscan and normal traffic.

A contribution to semantic description of images and videos: an application of soft biometrics

Humans have a high ability to extract visual data information acquired by sight. Trought a learning process, which starts at birth and continues throughout life, image interpretation becomes almost instinctively. At a glance, one can easily describe a scene with reasonable precision, naming its main components. Usually, this is done by extracting low-level features such as edges, shapes and textures, and associanting them to high level meanings. In this way, a semantic description of the scene is done. An example of this, is the human capacity to recognize and describe other people physical and behavioral characteristics, or biometrics. Soft-biometrics also represents inherent characteristics of human body and behaviour, but do not allow unique person identification. Computer vision area aims to develop methods capable of performing visual interpretation with performance similar to humans. This thesis aims to propose computer vison methods which allows high level information extraction from images in the form of soft biometrics. This problem is approached in two ways, unsupervised and supervised learning methods. The first seeks to group images via an automatic feature extraction learning , using both convolution techniques, evolutionary computing and clustering. In this approach employed images contains faces and people. Second approach employs convolutional neural networks, which have the ability to operate on raw images, learning both feature extraction and classification processes. Here, images are classified according to gender and clothes, divided into upper and lower parts of human body. First approach, when tested with different image datasets obtained an accuracy of approximately 80% for faces and non-faces and 70% for people and non-person. The second tested using images and videos, obtained an accuracy of about 70% for gender, 80% to the upper clothes and 90% to lower clothes. The results of these case studies, show that proposed methods are promising, allowing the realization of automatic high level information image annotation. This opens possibilities for development of applications in diverse areas such as content-based image and video search and automatica video survaillance, reducing human effort in the task of manual annotation and monitoring.

Implementación de un detector de coral utilizando filtros Gabor Wavelets y máquinas de aprendizaje

Este trabajo se enfoca en la implementación de un detector de arrecife de coral de desempeño rápido que se utiliza para un vehículo autónomo submarino (Autonomous Underwater Vehicle, AUV, por sus siglas en inglés). Una detección rápida de la presencia de coral asegura la estabilización del AUV frente al arrecife en el menor tiempo posible, evitando colisiones con el coral. La detección de coral se hace en una imagen que captura la escena que percibe la cámara del AUV. Se realiza una clasificación píxel por píxel entre dos clases: arrecife de coral y el plano de fondo que no es coral. A cada píxel de la imagen se le asigna un vector característico, el mismo que se genera mediante el uso de filtros Gabor Wavelets. Éstos son implementados en C++ y la librería OpenCV. Los vectores característicos son clasificados a través de nueve algoritmos de máquinas de aprendizaje. El desempeño de cada algoritmo se compara mediante la precisión y el tiempo de ejecución. El algoritmo de Árboles de Decisión resultó ser el más rápido y preciso de entre todos los algoritmos. Se creó una base de datos de 621 imágenes de corales de Belice (110 imágenes de entrenamiento y 511 imágenes de prueba).

SEQUENTIAL DECISIONS AND PREDICTIONS IN NATURAL LANGUAGE PROCESSING

Natural language processing has achieved great success in a wide range of ap- plications, producing both commercial language services and open-source language tools. However, most methods take a static or batch approach, assuming that the model has all information it needs and makes a one-time prediction. In this disser- tation, we study dynamic problems where the input comes in a sequence instead of all at once, and the output must be produced while the input is arriving. In these problems, predictions are often made based only on partial information. We see this dynamic setting in many real-time, interactive applications. These problems usually involve a trade-off between the amount of input received (cost) and the quality of the output prediction (accuracy). Therefore, the evaluation considers both objectives (e.g., plotting a Pareto curve). Our goal is to develop a formal understanding of sequential prediction and decision-making problems in natural language processing and to propose efficient solutions. Toward this end, we present meta-algorithms that take an existent batch model and produce a dynamic model to handle sequential inputs and outputs. Webuild our framework upon theories of Markov Decision Process (MDP), which allows learning to trade off competing objectives in a principled way. The main machine learning techniques we use are from imitation learning and reinforcement learning, and we advance current techniques to tackle problems arising in our settings. We evaluate our algorithm on a variety of applications, including dependency parsing, machine translation, and question answering. We show that our approach achieves a better cost-accuracy trade-off than the batch approach and heuristic-based decision- making approaches. We first propose a general framework for cost-sensitive prediction, where dif- ferent parts of the input come at different costs. We formulate a decision-making process that selects pieces of the input sequentially, and the selection is adaptive to each instance. Our approach is evaluated on both standard classification tasks and a structured prediction task (dependency parsing). We show that it achieves similar prediction quality to methods that use all input, while inducing a much smaller cost. Next, we extend the framework to problems where the input is revealed incremen- tally in a fixed order. We study two applications: simultaneous machine translation and quiz bowl (incremental text classification). We discuss challenges in this set- ting and show that adding domain knowledge eases the decision-making problem. A central theme throughout the chapters is an MDP formulation of a challenging problem with sequential input/output and trade-off decisions, accompanied by a learning algorithm that solves the MDP.

Search for Higgs Boson Production Beyond the Standard Model Using the Razor Kinematic Variables in pp Collisions at √s=8 TeV and Optimization of Higgs Boson Identification Using a Quantum Annealer

In the first part of this thesis we search for beyond the Standard Model physics through the search for anomalous production of the Higgs boson using the razor kinematic variables. We search for anomalous Higgs boson production using proton-proton collisions at center of mass energy √s=8 TeV collected by the Compact Muon Solenoid experiment at the Large Hadron Collider corresponding to an integrated luminosity of 19.8 fb^-1.

In the second part we present a novel method for using a quantum annealer to train a classifier to recognize events containing a Higgs boson decaying to two photons. We train that classifier using simulated proton-proton collisions at √s=8 TeV producing either a Standard Model Higgs boson decaying to two photons or a non-resonant Standard Model process that produces a two photon final state.

The production mechanisms of the Higgs boson are precisely predicted by the Standard Model based on its association with the mechanism of electroweak symmetry breaking. We measure the yield of Higgs bosons decaying to two photons in kinematic regions predicted to have very little contribution from a Standard Model Higgs boson and search for an excess of events, which would be evidence of either non-standard production or non-standard properties of the Higgs boson. We divide the events into disjoint categories based on kinematic properties and the presence of additional b-quarks produced in the collisions. In each of these disjoint categories, we use the razor kinematic variables to characterize events with topological configurations incompatible with typical configurations found from standard model production of the Higgs boson.

We observe an excess of events with di-photon invariant mass compatible with the Higgs boson mass and localized in a small region of the razor plane. We observe 5 events with a predicted background of 0.54 ± 0.28, which observation has a p-value of 10^-3 and a local significance of 3.35σ. This background prediction comes from 0.48 predicted non-resonant background events and 0.07 predicted SM higgs boson events. We proceed to investigate the properties of this excess, finding that it provides a very compelling peak in the di-photon invariant mass distribution and is physically separated in the razor plane from predicted background. Using another method of measuring the background and significance of the excess, we find a 2.5σ deviation from the Standard Model hypothesis over a broader range of the razor plane.

In the second part of the thesis we transform the problem of training a classifier to distinguish events with a Higgs boson decaying to two photons from events with other sources of photon pairs into the Hamiltonian of a spin system, the ground state of which is the best classifier. We then use a quantum annealer to find the ground state of this Hamiltonian and train the classifier. We find that we are able to do this successfully in less than 400 annealing runs for a problem of median difficulty at the largest problem size considered. The networks trained in this manner exhibit good classification performance, competitive with the more complicated machine learning techniques, and are highly resistant to overtraining. We also find that the nature of the training gives access to additional solutions that can be used to improve the classification performance by up to 1.2% in some regions.

Online Resources for Improving EFL Teaching and Learning

El presente artículo plantea el beneficio de utilizar recursos en línea en la enseñanza del inglés. Destaca que los estudiantes fortalecerán, no solo el uso de la lengua meta, sino también el de los recursos tecnológicos. Para ejemplificar, se presenta una serie de ejercicios en línea que desarrollan diversas habilidades de la lengua meta como también las ventajas y desventajas del uso de los mismos. Por último, se comparten los resultados obtenidos de una encuesta aplicada sobre el uso de recursos en línea en las clases de inglés.The benefit of using online sources in the EFL class is analyzed here starting from the perspective that this helps students improve not only their use of the language but also their use of technology. Sample online exercises focusing on the development of different language skills are described here, along with the advantages and disadvantages of using online sources. Finally, the results obtained from a survey on the use of online sources in the EFL classes are presented.