Neural network approaches have become popular in the field of automatic speech recognition (ASR). Most ASR methods use audio data to classify words. Lip reading ASR techniques utilize only video data, which compensates for noisy environments where audio may be compromised. A comprehensive approach, including the vetting of datasets and development of a preprocessing chain, to video-based ASR is developed. This approach will be based on neural networks, namely 3D convolutional neural networks (3D-CNN) and Long short-term memory (LSTM). These types of neural networks are designed to take in temporal data such as videos. Various combinations of different neural network architecture and preprocessing techniques are explored. The best performing neural network architecture, a CNN with bidirectional LSTM, compares favorably against recent works on video-based ASR.

Safe and secure transportation of hazardous materials (hazmat) is a challenging issue in terms of optimizing risk to society and simultaneously making the shipment delivery economical. The most important safety concern of hazardous material transportation is accidents causing multiple causalities. The potential risk to society from hazmat transportation has led to the evolution of a new threat from terrorism. Malicious entities can turn hazmat vehicles into weapons causing explosions in high profile locations. The present research is divided into two parts. First, a neural network model is developed to identify when a hazmat truck deviates from its pre-specified path based on its location in the road network. The model identifies abnormal diversions in hazmat carriers' paths considering normal diversions arising due to incidents. The second part of this thesis develops a methodology for predicting different paths that could be taken by malicious entities heading towards a target after successfully hijacking a hazmat vehicle. The path prediction methodology and the neural network methodology are implemented on the network between Baltimore, Maryland and Washington, DC. The trained neural network model classified nodes in the network with a satisfactory performance .The path prediction algorithm was used to calculate the paths to two targets located at the International Dulles Airport and the National Mall in Washington, DC. Based on this research, the neural network methodology is a promising technology for detecting a hijacked vehicle in its initial stages of diversion from its pre-specified path. Possible paths to potential targets are plotted and points of overlap among paths are identified. Overlaps are critical locations where extra security measures can be taken for preventing destruction. Thus, integrating both models gives a comprehensive methodology for detecting the initial diversion and then predicting the possible paths of malicious entities towards targets and could provide an important tool for law enforcement agencies minimizing catastrophic events. / Master of Science

The expectations on embedded systems have grown incredibly in recent years. Not only are there more applications for them than ever, the applications are increasingly complex, and their security is essential. To meet such demanding goals, designers and programmers are always looking for more efficient methods of computation. One technique that has gained attention over the past couple of decades is bitsliced software. In addition to high efficiency in certain situations, including block ciphers computation, it has been used in designs to resist hardware attacks. However, this technique requires both program and data to be in a specific format. This requirement makes writing bitsliced software by hand laborious and adds computational overhead to transpose the data before and after computation. This work describes a code generation tool that produces it from a higher-level description in Verilog. By supporting the synthesis of sequential circuits, this tool extends bitsliced software to parallel synchronous software. This tool is then used to implement a method for accelerating software neural network processing with reduced-precision computation on highly constrained devices. To address the data transposition overhead and to support a hardware attack-resistant architecture, a custom DMA controller is introduced that efficiently transposes the data as it transfers along with dedicated hardware for masking and redundancy generation. In combination, these tools make bitsliced software and its benefits more accessible to system designers and programmers. / Master of Science / Small computers embedded in devices, such as cars, smart devices, and other electronics, face many challenges. Often, they are pushed to their limits by designers and programmers to reach acceptable levels of performance. The increasing complexity of the applications they run compounds with the need for these applications to be secure. The programmers are always looking for better, more efficient methods of doing computations. Over the past two decades bitsliced software has gained attention as a technique that can, in certain situations, be more efficient than standard software. It also has properties that make it useful for designs implementing secure software. However, writing bitsliced software by hand is a laborious task, and the data input to the software needs to be in a specific format. To make writing the software easier, a tool that generates it from the well-known Verilog hardware description language is discussed in this work. This tool is then used to implement a method to accelerate artificial intelligence calculations on highly constrained computers. A custom hardware module is also introduced to speed up the formatting of data for bitsliced processing. In combination, these tools make bitsliced software and its benefits more accessible.

Yes / Corneal images can be acquired using confocal microscopes which provide detailed images of the different layers inside the cornea. Most corneal problems and diseases occur in one or more of the main corneal layers: the epithelium, stroma and endothelium. Consequently, for automatically extracting clinical information associated with corneal diseases, or evaluating the normal cornea, it is important also to be able to automatically recognise these layers easily. Artificial intelligence (AI) approaches can provide improved accuracy over the conventional processing techniques and save a useful amount of time over the manual analysis time required by clinical experts. Artificial neural networks (ANN) and adaptive neuro fuzzy inference systems (ANFIS), are powerful AI techniques, which have the capability to accurately classify the main layers of the cornea. The use of an ANFIS approach to analyse corneal layers is described for the first time in this paper, and statistical features have been also employed in the identification of the corneal abnormality. An ANN approach is then added to form a combined committee machine with improved performance which achieves an accuracy of 100% for some classes in the processed data sets. Three normal data sets of whole corneas, comprising a total of 356 images, and seven abnormal corneal images associated with diseases have been investigated in the proposed system. The resulting system is able to pre-process (quality enhancement, noise removal), classify (whole data sets, not just samples of the images as mentioned in the previous studies), and identify abnormalities in the analysed data sets. The system output is visually mapped and the main corneal layers are displayed. 3D volume visualisation for the processed corneal images as well as for each individual corneal cell is also achieved through this system. Corneal clinicians have verified and approved the clinical usefulness of the developed system especially in terms of underpinning the expertise of ophthalmologists and its applicability in patient care.

The paper discusses how to boost corporate value systematically based on the theory of the five elements in traditional Chinese culture; firstly, the categorization and analogy approach from the theory of the five elements is applied to make the five elements correspond to the five major factors influencing corporate value: metal, wood, water, fire, and earth, correspond to business innovation, financial capability, operational efficiency, competitive barriers, and comprehensive capacity, respectively, and the logical relation in these five elements is described and explained according to the principle of inter-promoting relation in the five elements. Continuous business innovation will bring about the improvement of operational efficiency, which in turn will be directly reflected in the growth of financial indicators, thereby providing an enterprise with the performance growth and cash flow guarantee; depending on a better profitability, the enterprise is capable of attracting and accumulating resources such as talents and capital, gaining competitive advantages and forming competitive barriers, as well as promoting the improvement of comprehensive capabilities and the cultivation of corporate culture; in addition, the improvement of corporate atmosphere can promote innovation more effectively, and thus boost corporate value systematically. Moreover, inspired by neural networks, the paper establishes an umbrella corporate value scoring system by combining it with the M-P neuron model, which is composed of the umbrella chart of corporate value scoring and the five-element neural network evaluation model. With respect to an enterprise, the five factors can be scored after an analysis is performed on its financial reports, industry rankings, corporate announcements and other information, and a five-dimensional umbrella radar chart can clearly present the performance of each factor influencing its value; the industry average is taken into account in scoring, the umbrella chart of each enterprise is not isolated but comparable because they all establish a certain connection with the industry; the five-element neural network evaluation model is regarded as a tool to facilitate enterprises to calculate the corporate value score, also with the help of the initial scoring of each element. The score reflects the inter-promoting relation principle of the five elements and can show how much each element can contribute to the improvement of the corporate value under the mutual influence, thereby highlighting the concept of systematic boost of corporate value. In this paper, the five elements analysis theory is applied to analyze the five enterprises: Ali Health, Wal-Mart, Tesla, Kweichow Moutai, and Haidilao, and find out the advantages of each company by making use of the inter-promoting relationship between the promoting elements, promoted elements and advantageous elements, and conclude their shortcomings of the companies and provide development suggestions. Finally, this paper explores the application method of the principle of inter-restricting in case analysis, i.e. developing the elements restricting advantageous elements, and promoting the systematic boost of corporate value from the perspective of balance. This chapter also expands the theory of the five elements from a company to the whole industry, and established a three-level classification system of "industry-segment -company". Each company in industry has its own advantages and it is unavailable to extract the advantageous element in the whole industry. However, in the process of dividing the industry into multiple segments, the present research concludes that the leading enterprises in each segment represent the future development trend of the field to a certain extent. The advantages of this enterprise tend to represent the advantages of this field. As a result, the leading enterprises can obtain various high-quality resources in the industry, which is helpful to further boost their corporate value. / Business Administration/Finance

京都大学 / 新制・課程博士 / 博士(理学) / 甲第24174号 / 理博第4865号 / 新制||理||1696(附属図書館) / 京都大学大学院理学研究科地球惑星科学専攻 / (主査)准教授 成瀬 元, 准教授 堤 昭人, 教授 野口 高明 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

<p> Connectionist models of language acquisition typically have difficulty with systematicity, or the ability for the network to generalize its limited experience with language to novel utterances. In this way, connectionist systems learning grammar from a set of example sentences tend to store a set of specific instances, rather than a generalized abstract knowledge of the process of grammatical combination. Further, recent models that do show limited systematicity do so at the expense of simultaneously storing explicit lexical knowledge, and also make use of both developmentally-implausible training data and biologically-implausible learning rules. Consequently, this research program develops a novel unsupervised neural network architecture, and applies this architecture to the problem of systematicity in language models.</p> <p> In the first of several studies, a connectionist architecture capable of simultaneously storing explicit and separate representations of both conceptual and grammatical information is developed, where this architecture is a hybrid of both a self-organizing map and an intra-layer Hebbian associative network. Over the course of several studies, this architecture's capacity to acquire linguistic grammar is evaluated, where the architecture is progressively refined until it is capable of acquiring a benchmark grammar consisting of several difficult clausal sentence structures - though it must acquire this grammar at the level of grammatical category, rather than the lexical level.</p> <p> The final study bridges the gap between the lexical and grammatical category levels, and develops an activation function based on a semantic feature co-occurrence metric. In concert with developmentally-plausible sensorimotor grounded conceptual representations, it is shown that a network using this metric is able to undertake a process of semantic bootstrapping, and successfully acquire separate explicit representations at the level of the concept, part-of-speech category, and grammatical sequence. This network demonstrates broadly systematic behaviour on a difficult test of systematicity, and extends its knowledge of grammar to novel sensorimotor-grounded words.</p> / Thesis / Doctor of Philosophy (PhD)

UAV (Unmanned Airborne Vehicle) is a source of entertainment and a pleasurable experience, attracting many young people to pursue it as a hobby. With the potential increase in the number of UAVs, the risk of using them for malicious purposes also increases. In addition, birds and UAVs have very similar maneuvers during flights. These UAVs can also carry a significant payload, which can have unintended consequences. Therefore, detecting UAVs near red-zone areas is an important problem. In addition, small UAVs can record video from large distances without being spotted by the naked eye. An appropriate network of sensors may be needed to foresee the arrival of such entities from a safe distance before they pose any danger to the surrounding areas. Despite the growing interest in UAV detection, limited research has been conducted in this area due to a lack of available data for model training. This thesis proposes a novel approach to address this challenge by leveraging experimental data collected in real-time using high-sensitivity sensors instead of relying solely on simulations. This approach allows for improved model accuracy and a better representation of the complex and dynamic environments in which UAVs operate, which are difficult to simulate accurately. The thesis further explores the application of machine learning and sensor fusion algorithms to detect UAVs and distinguish them from other objects, such as birds, in real-time. Specifically, the thesis utilizes YOLOv3 with deep sort and sensor fusion algorithms to achieve accurate UAV detection. In this study, we employed YOLOv3, a deep learning model known for its high efficiency and complexity, to facilitate real-time drone versus bird detection. To further enhance the reliability of the system, we incorporated sensor fusion, leading to a more stable and accurate real-time system, and mitigating the incidence of false detections. Our study indicates that the YOLOv3 model outperformed the state-of-the-art models in terms of both speed and robustness, achieving a high level of confidence with a score above 95%. Moreover, the YOLOv3 model demonstrated a promising capability in real-time drone versus bird detection, which suggests its potential for practical applications

Wildfires burn millions of acres of land each year leading to the destruction of homes and wildland ecosystems while costing governments billions in funding. As climate change intensifies drought volatility across the Western United States, wildfires are likely to become increasingly severe. Wildfire risk assessment and hazard maps are currently employed by fire services, but can often be outdated. This paper introduces an image-based dataset using climate and wildfire data from NASA’s Moderate Resolution Imaging Spectroradiometer (MODIS). The dataset consists of 32 climate and topographical layers captured across 0.1 deg by 0.1 deg tiled regions in California and Nevada between 2015 and 2020, associated with whether the region later saw a wildfire incident. We trained a convolutional neural network (CNN) with the generated dataset to predict whether a region will see a wildfire incident given the climate data of that region. Convolutional neural networks are able to find spatial patterns in their multi-dimensional inputs, providing an additional layer of inference when compared to logistic regression (LR) or artificial neural network (ANN) models. To further understand feature importance, we performed an ablation study, concluding that vegetation products, fire history, water content, and evapotranspiration products resulted in increases in model performance, while land information products did not. While the novel convolutional neural network model did not show a large improvement over previous models, it retained the highest holistic measures such as area under the curve and average precision, indicating it is still a strong competitor to existing models. This introduction of the convolutional neural network approach expands the wealth of knowledge for the prediction of wildfire incidents and proves the usefulness of the novel, image-based dataset.

The use of orthogonal frequency-division multiplexing (OFDM) by wireless standards is often preferred due to its high spectral efficiency and ease of implementation. However, data transmission via OFDM still suffers when passing through a noisy channel. In order to maximize the abilities of OFDM, channel effects must be corrected. Unfortunately, channel estimation is often difficult due to the nonlinearity and randomness present in a practical communication channel. Recently, machine learning based approaches have been used to improve existing channel estimation algorithms for a more efficient transmission. This thesis investigates the application of artificial neural networks (ANNs) as a means of improving existing channel estimation techniques. Multi-layer feed forward neural networks (FNNs) and convolutional neural networks (CNNs) are tested on a variety of random fading channels with different signal-to-noise ratios (SNRs) via computer simulations. Compared to the conventional least squares (LS) algorithm, the approach based on CNN can reduce the bit error rate (BER) of data transmission by an average of 47.59%.