abstract: Automated planning addresses the problem of generating a sequence of actions that enable a set of agents to achieve their goals.This work investigates two important topics from the field of automated planning, namely model-lite planning and multi-agent planning. For model-lite planning, I focus on a prominent model named Annotated PDDL and it's related application of robust planning. For this model, I try to identify a method of leveraging additional domain information (available in the form of successful plan traces). I use this information to refine the set of possible domains to generate more robust plans (as compared to the original planner) for any given problem. This method also provides us a way of overcoming one of the major drawbacks of the original approach, namely the need for a domain writer to explicitly identify the annotations. For the second topic, the central question I ask is ``{\em under what conditions are multiple agents actually needed to solve a given planning problem?}''. To answer this question, the multi-agent planning (MAP) problem is classified into several sub-classes and I identify the conditions in each of these sub-classes that can lead to required cooperation (RC). I also identify certain sub-classes of multi-agent planning problems (named DVC-RC problems), where the problems can be simplified using a single virtual agent. This insight is later used to propose a new planner designed to solve problems from these subclasses. Evaluation of this new planner on all the current multi-agent planning benchmarks reveals that most current multi-agent planning benchmarks only belong to a small subset of possible classes of multi-agent planning problems. / Dissertation/Thesis / Masters Thesis Computer Science 2016

abstract: Achieving human level intelligence is a long-term goal for many Artificial Intelligence (AI) researchers. Recent developments in combining deep learning and reinforcement learning helped us to move a step forward in achieving this goal. Reinforcement learning using a delayed reward mechanism is an approach to machine intelligence which studies decision making with control and how a decision making agent can learn to act optimally in an environment-unaware conditions. Q-learning is one of the model-free reinforcement directed learning strategies which uses temporal differences to estimate the performances of state-action pairs called Q values. A simple implementation of Q-learning algorithm can be done using a Q table memory to store and update the Q values. However, with an increase in state space data due to a complex environment, and with an increase in possible number of actions an agent can perform, Q table reaches its space limit and would be difficult to scale well. Q-learning with neural networks eliminates the use of Q table by approximating the Q function using neural networks. Autonomous agents need to develop cognitive properties and become self-adaptive to be deployable in any environment. Reinforcement learning with Q-learning have been very efficient in solving such problems. However, embedded systems like space rovers and autonomous robots rarely implement such techniques due to the constraints faced like processing power, chip area, convergence rate and cost of the chip. These problems present a need for a portable, low power, area efficient hardware accelerator to accelerate the process of such learning. This problem is targeted by implementing a hardware schematic architecture for Q-learning using Artificial Neural networks. This architecture exploits the massive parallelism provided by neural network with a dedicated fine grain parallelism provided by a Field Programmable Gate Array (FPGA) thereby processing the Q values at a high throughput. Mars exploration rovers currently use Xilinx-Space-grade FPGA devices for image processing, pyrotechnic operation control and obstacle avoidance. The hardware resource consumption for the architecture has been synthesized considering Xilinx Virtex7 FPGA as the target device. / Dissertation/Thesis / Masters Thesis Engineering 2016

In this thesis we address the problem of adaptive control in complex stochastic systems when the system parameters are both known and unknown. The type of models we consider are those which, in the full information case, are known as Markov Decision Processes. We introduce versions of two new algorithms, the optimiser and the p-learner. The optimiser is a simulation based method for finding optimal values and optimal policies when the system parameters are known. The p-learner is an algorithm for learning about the state transition probabilities; we use it in conjunction with the optimiser when the system parameters are unknown. We carefully discuss the choice of different components in the different versions of the algorithms, and we look at two extended case studies to evaluate their performances over a range of different learning parameters. In each case, we compare the results with that of a deterministic method. We also address the convergence of the solutions generated by the optimiser

abstract: There has been a lot of research in the field of artificial intelligence about thinking machines. Alan Turing proposed a test to observe a machine's intelligent behaviour with respect to natural language conversation. The Winograd schema challenge is suggested as an alternative, to the Turing test. It needs inferencing capabilities, reasoning abilities and background knowledge to get the answer right. It involves a coreference resolution task in which a machine is given a sentence containing a situation which involves two entities, one pronoun and some more information about the situation and the machine has to come up with the right resolution of a pronoun to one of the entities. The complexity of the task is increased with the fact that the Winograd sentences are not constrained by one domain or specific sentence structure and it also contains a lot of human proper names. This modification makes the task of association of entities, to one particular word in the sentence, to derive the answer, difficult. I have developed a pronoun resolver system for the confined domain Winograd sentences. I have developed a classifier or filter which takes input sentences and decides to accept or reject them based on a particular criteria. Once the sentence is accepted. I run parsers on it to obtain the detailed analysis. Furthermore I have developed four answering modules which use world knowledge and inferencing mechanisms to try and resolve the pronoun. The four techniques I use are : ConceptNet knowledgebase, Search engine pattern counts,Narrative event chains and sentiment analysis. I have developed a particular aggregation mechanism for the answers from these modules to arrive at a final answer. I have used caching technique for the association relations that I obtain for different modules, so as to boost the performance. I run my system on the standard ‘nyu dataset’ of Winograd sentences and questions. This dataset is then restricted, by my classifier, to 90 sentences. I evaluate my system on this 90 sentence dataset. When I compare my results against the state of the art system on the same dataset, I get nearly 4.5 % improvement in the restricted domain. / Dissertation/Thesis / M.S. Computer Science 2013

<p> Social dynamics is concerned with the interactions of individuals and the resulting group behaviors. It models the temporal evolution of social systems via the interactions of the individuals within these systems. The availability of large-scale data in social networks and sensor networks offers an unprecedented opportunity to predict state changing events at the individual level. Examples of such events are disease infection, rumor propagation and opinion transition in elections, etc. Unlike previous research focusing on the collective effects of social systems, we want to make efficient inferences on the individual level.</p><p> Two main challenges are addressed: temporal modeling and computational complexity. The interaction pattern for each individual keeps changing over the time, i.e., an individual interacts with different individuals at different times. Second, as the number of tracked individual increases, the computational complexity grows exponentially with traditional sequential data analysis. </p><p> The contributions are: (i) leverage social networks and sensor networks data to make tractable inferences on both individual behaviors and collective effects in social dynamics. (ii) use the stochastic kinetic model to summarize dynamic interactions among individuals and simplify the state transition probabilities. (iii) propose an efficient variational inference algorithm whose complexity grows <i>linearly</i> with the number of tracked individuals <i> M</i>. Given the state space <i>K</i> of a single individual and the total number of time steps <i>T</i>, the complexity of naive brute-force approach is <i>O(K^MT)</i> and the complexity of existing exact inference approach is <i>O(K^MT)</i>. In comparison, the complexity of the proposed algorithm is<i> O(K²MT)</i>. In practice, it requires several iterations to converge. </p><p> In the empirical study concerning epidemics dynamics, given wireless sensor network data collected from more than ten thousand people (M = 13,888) over three years (T = 3465), we use the proposed algorithm to track disease transmission, and predict the probability of infection for each individual (K = 2) along the time until convergence (I=5). It is more efficient than state of the art sampling methods, i.e., MCMC and particle filter, while achieving high accuracy.</p>

No description available.

Sentiment Analysis consists of recognizing sentiment orientation towards specific subjects within natural language texts. Most research in this area focuses on classifying documents as positive or negative. The purpose of this thesis is to quantitatively evaluate subjective opinions of customer reviews using a five star rating system, which is widely used on on-line review web sites, and to try to make the predicted score as accurate as possible. Firstly, this thesis presents two methods for rating reviews: classifying reviews by supervised learning methods as multi-class classification does, or rating reviews by using association scores of sentiment terms with a set of seed words extracted from the corpus, i.e. the unsupervised learning method. We extend the feature selection approach used in Turney's PMI-IR estimation by introducing semantic relatedness measures based up on the content of WordNet. This thesis reports on experiments using the two methods mentioned above for rating reviews using the combined feature set enriched with WordNet-selected sentiment terms. The results of these experiments suggest ways in which incorporating WordNet relatedness measures into feature selection may yield improvement over classification and unsupervised learning methods which do not use it. Furthermore, via ordinal meta-classifiers, we utilize the ordering information contained in the scores of bank reviews to improve the performance, we explore the effectiveness of re-sampling for reducing the problem of skewed data, and we check whether discretization benefits the ordinal meta-learning process. Finally, we combine the unsupervised and supervised meta-learning methods to optimize performance on our sentiment prediction task.

This multiple case study describes students who are highly emotionally competent yet have learning disabilities. The study sheds light on how such students perceive their educational experience and begins to answer inter-related questions, such as how emotional strengths assist with learning disabilities. A multiple case study design was used. The participant group ranged from 11 to 16 years of age and came from two separate schools which actively work with students diagnosed with learning disabilities. The study was divided into two phases. In the first phase, the Mayer—Salovey—Caruso Emotional Intelligence Test-Youth Version (MSCEIT-YV) was given to students in the two participating classes. The two students from each class who achieved the highest scores on the MSCEIT-YV were then asked to participate in the second phase of the study. Here, the researcher conducted observations of the participants within the school environment. Additionally, the participants attended a semi-structured interview, with interview questions based on the MSCEIT-YV and school related scenarios. Themes that emerged were then analyzed and compared within and between cases as well as with emotional intelligence research. Case study descriptions emerged from this analysis and a brief follow up interview was conducted with one family member and the participating student as a means of sharing and verifying findings. Participants revealed varying ability with emotional intelligence. However, all students demonstrated strong abilities with the ‘Strategic Emotional Reasoning’ Skills associated with Mayer, Salovey and Caruso’s (2004) theory of emotional intelligence. Moreover, all students showed a strong ability to use their emotional intelligence to improve academic functioning, with one student in particular displaying outstanding abilities and insights into emotional intelligence. The study contributes to our understanding of the complexity of ability and disability that can exist within students diagnosed with learning disabilities; this understanding, in turn, may be reflected in how these students are perceived and understood by researchers and teachers alike. / Education, Faculty of / Educational and Counselling Psychology, and Special Education (ECPS), Department of / Graduate

<p> Automatic age estimation from real-world and wild face images is a challenging task and has an increasing importance due to its wide range of applications in current and future lifestyles. As a result of increasing age specific human-computer interactions, it is expected that computerized systems should be capable of estimating the age from face images and respond accordingly. Over the past decade, many research studies have been conducted on automatic age estimation from face images. </p><p> In this research, new approaches for enhancing age classification of a person from face images based on deep neural networks (DNNs) are proposed. The work shows that pre-trained CNNs which were trained on large benchmarks for different purposes can be retrained and fine-tuned for age estimation from unconstrained face images. Furthermore, an algorithm to reduce the dimension of the output of the last convolutional layer in pre-trained CNNs to improve the performance is developed. Moreover, two new jointly fine-tuned DNNs frameworks are proposed. The first framework fine-tunes tow DNNs with two different feature sets based on the element-wise summation of their last hidden layer outputs. While the second framework fine-tunes two DNNs based on a new cost function. For both frameworks, each has two DNNs, the first DNN is trained by using facial appearance features that are extracted by a well-trained model on face recognition, while the second DNN is trained on features that are based on the superpixels depth and their relationships. </p><p> Furthermore, a new method for selecting robust features based on the power of DNN and <i>l₂₁-norm</i> is proposed. This method is mainly based on a new cost function relating the DNN and the L21 norm in one unified framework. To learn and train this unified framework, the analysis and the proof for the convergence of the new objective function to solve minimization problem are studied. Finally, the performance of the proposed jointly fine-tuned networks and the proposed robust features are used to improve the age estimation from the facial images. The facial features concatenated with their corresponding robust features are fed to the first part of both networks and the superpixels features concatenated with their robust features are fed to the second part of the network </p><p> Experimental results on a public database show the effectiveness of the proposed methods and achieved the state-of-art performance on a public database. </p><p>

A methodology for knowledge acquisition from terminology databases is presented. The methodology outlines how the content of a terminology database can be mapped onto a knowledge base with a minimum of human intervention. Typically, terms are defined and elaborated by terminologists by using sentences that have a common syntactic and semantic structure. It has been argued that in defining terms, terminologists use a local grammar and that this local grammar can be used to parse the definitions. The methodology has been implemented in a program called DEARSys (Definition Analysis and Representation System), that reads definition sentences and extracts new concepts and conceptual relations about the defined terms. The linguistic component of the system is a parser for the sublanguage of terminology definitions that analyses a definition into its logical form, which in turn is mapped onto a frame-based representation. The logical form is based on first-order logic (FOL) extended with untyped lambda calculus. Our approach is data-driven and domain independent; it has been applied to definitions of various domains. Experiments were conducted with human subjects to evaluate the information acquired by the system. The results of the preliminary evaluation were encouraging.