No description available.

No description available.

[pt] Detectar reservatórios de hidrocarbonetos a partir de um levantamento sísmico é uma tarefa complexa, que requer profissionais especializados e muito tempo de trabalho. Por isso, atualmente, existem muitas pesquisas que buscam automatizar essa tarefa utilizando redes neurais profundas. Seguindo o sucesso das redes convolucionais profundas, CNNs, na identificação de objetos em imagens e vídeos, as CNNs tem sido utilizadas como detectores de eventos geológicos nas imagens sísmica. O treinamento de uma rede neural profunda atual, entretanto, requer centenas de milhares de dados rotulados. Se tratarmos os dados sísmicos como imagens, os reservatórios de hidrocarbonetos geralmente constituem uma pequena sub imagem incapaz de fornecer tantas amostras. A metodologia proposta nesta dissertação trata o dado sísmico como um conjunto de traços e a amostra que alimenta a rede neural são trechos de um sinal unidimensional parecido com um sinal de som ou voz. Com essa entrada uma marcação de um reservatório numa sísmica geralmente já fornece o número necessário de amostras rotuladas para o treinamento. Um outro aspecto importante da nossa proposta é a utilização de uma rede neural recorrente. A influencia de um reservatório de hidrocarboneto num traço sísmico se dá não somente no local onde ele se encontra, mas em todo o traço que se segue. Por isso propomos a utilização de uma rede do tipo longa memória de curto prazo (Long Short-Term Memory, LSTM) para caracterizar regiões que apresentem assinaturas de gás em imagens sísmicas. Esta dissertação detalha ainda a implementação da metodologia proposta e os testes feitos nos dados sísmicos públicos Netherlands F3-Block. Os resultados alcançados avaliados pelos índices de sensibilidade, especificidade, acurácia e AUC foram todos excelentes, acima de 95 por cento. / [en] Detecting hydrocarbon reservoirs from a seismic survey is a complex task, requiring specialized professionals and long time. Consequently, many authors today seek to automate this task by using deep neural networks. Following the success of deep convolutional networks, CNNs, in the identification of objects in images and videos, CNNs have been used as detectors of geological events in seismic images. Training a deep neural network, however, requires hundreds of thousands of labeled data, that is, samples that we know the response that the network must provide. If we treat seismic data as images, the hydrocarbon reservoirs usually constitute a small sub-image unable to provide so many samples. The methodology proposed in this dissertation treats the seismic data as a set of traces and the sample that feeds the neural network are fragments of a onedimensional signal resembling a sound or voice signal. A labeled reservoir seismic image usually provides the required number of labeled one-dimensional samples for training. Another important aspect of our proposal is the use of a recurrent neural network. The influence of a hydrocarbon reservoir on a seismic trace occurs not only in its location but throughout the trace that follows. For this reason, we propose the use of a Long Short-Term Memory, LSTM, network to characterize regions that present gas signatures in seismic images. This dissertation further details the implementation of the proposed methodology and test results on the Netherlands F3-Block public seismic data. The results on this data set, evaluated by sensitivity, specificity, accuracy and AUC indexes, are all excellent, above 95 percent.

<p> This thesis presents a novel hybrid methodology using Recurrent Neural Network and Dynamic Time Warping to solve the mode estimation problem of a radar warning receiver (RWR). The RWR is an electronic support (ES) system with the primary objective to estimate the threat posed by an unfriendly (hostile) radar in an electronic warfare (EW) environment. One such radar is the multi-function radar (MFR), which employs complex signal architecture to perform multiple tasks. As the threat posed by the radar directly depends on its current mode of operation, it is vital to estimate and track the mode of the radar. The proposed method uses a recurrent neural network (echo state network and recurrent multi-layer perceptron) trained in a supervised manner, with the dynamic time warping algorithm as the post processor to estimate the mode of operation. A grid filter in Bayesian framework is then applied to the dynamic time warp estimate to provide an accurate posterior estimate of the operational mode of the MFR. This novel approach is tested on an EW scenario via simulation by employing a hypothetical MFR. Based on the simulation results, we conclude that the hybrid echo state network is more suitable than its recurrent multi-layer perceptron counterpart for the mode estimation problem of a RWR.</p> / Thesis / Master of Applied Science (MASc)

There is a long history in Kiruna of conducting research on the physics of the aurora borealis. There is also a long history of providing tourists with great opportunities to see the auroras. Planning such tourist activities can be challenging since the auroras are hard to predict. Reliable forecasts would be a valuable tool for researchers as well as for tourists and tour guides. One tool that is already available for both researchers and tourists is the all-sky camera in Kiruna, which is operated by the Swedish Institute of Space Physics (IRF). There has been a digital all-sky camera in operation in Kiruna for over 20 years. From the images captured by this camera, the IRF has developed a numerical index - the auroral index.  Forecasting time series with neural network algorithms is a well studies subject. There are many examples from a wide range of felds, including space weather. A type of neural network that has often been successfully used for time series forecasting is the Recurrent Neural Network (RNN), and more specifcally the Long short-term memory (LSTM).  This thesis evaluates the auroral index - in combination with data from the solar wind - as training data for recurrent neural networks. Furthermore, it attempts to fnd a LSTM neural network model capable of making reliable forecasts of the auroral index. The Keras and TensorFlow software libraries are used to build and train the neural network model. Some challenges with the auroral index - when utilized as training data for neural networks - are identifed. The produced LSTM neural network models are not accurate enough for deployment as a production level service. Further development might improve on this. Finally, this thesis suggests future work that may contribute to better forecasting models for auroras in the Kiruna region.

Recurrent neural networks (RNNs) have been widely used for supervised pattern recognition and exploring the underlying spatio-temporal correlation. However, due to the vanishing/exploding gradient problem, training a fully connected RNN in many cases is very difficult or even impossible. The difficulties of training traditional RNNs, led us to reservoir computing (RC) which recently attracted a lot of attention due to its simple training methods and fixed weights at its recurrent layer. There are three different categories of RC systems, namely, echo state networks (ESNs), liquid state machines (LSMs), and delayed feedback reservoirs (DFRs). In this dissertation a novel structure of RNNs which is inspired by dynamic delayed feedback loops is introduced. In the reservoir (recurrent) layer of DFR, only one neuron is required which makes DFRs extremely suitable for hardware implementations. The main motivation of this dissertation is to introduce an energy efficient, and easy to train RNN while this model achieves high performances in different tasks compared to the state-of-the-art. To improve the energy efficiency of our model, we propose to adopt spiking neurons as the information processing unit of DFR. Spiking neural networks (SNNs) are the most biologically plausible and energy efficient class of artificial neural networks (ANNs). The traditional analog ANNs have marginal similarity with the brain-like information processing. It is clear that the biological neurons communicate together through spikes. Therefore, artificial SNNs have been introduced to mimic the biological neurons. On the other hand, the hardware implementation of SNNs have shown to be extremely energy efficient. Towards achieving this overarching goal, this dissertation presents a spiking DFR (SDFR) with novel encoding schemes, and defense mechanisms against adversarial attacks. To verify the effectiveness and performance of the SDFR, it is adopted in three different applications where there exists a significant Spatio-temporal correlations. These three applications are attack detection in smart grids, spectrum sensing of multi-input-multi-output(MIMO)-orthogonal frequency division multiplexing (OFDM) Dynamic Spectrum Sharing (DSS) systems, and video-based face recognition. In this dissertation, the performance of SDFR is first verified in cyber attack detection in Smart grids. Smart grids are a new generation of power grids which guarantee a more reliable and efficient transmission and delivery of power to the costumers. A more reliable and efficient power generation and distribution can be realized through the integration of internet, telecommunication, and energy technologies. The convergence of different technologies, brings up opportunities, but the challenges are also inevitable. One of the major challenges that pose threat to the smart grids is cyber-attacks. A novel method is developed to detect false data injection (FDI) attacks in smart grids. The second novel application of SDFR is the spectrum sensing of MIMO-OFDM DSS systems. DSS is being implemented in the fifth generation of wireless communication systems (5G) to improve the spectrum efficiency. In a MIMO-OFDM system, not all the subcarriers are utilized simultaneously by the primary user (PU). Therefore, it is essential to sense the idle frequency bands and assign them to the secondary user (SU). The effectiveness of SDFR in capturing the spatio-temporal correlation of MIMO-OFDM time-series and predicting the availability of frequency bands in the future time slots is studied as well. In the third application, the SDFR is modified to be adopted in video-based face recognition. In this task, the SDFR is leveraged to recognize the identities of different subjects while they rotate their heads in different angles. Another contribution of this dissertation is to propose a novel encoding scheme of spiking neurons which is inspired by the cognitive studies of rats. For the first time, the multiplexing of multiple neural codes is introduced and it is shown that the robustness and resilience of the spiking neurons is increased against noisy data, and adversarial attacks, respectively. Adversarial attacks are small and imperceptible perturbations of the input data, which have shown to be able to fool deep learning (DL) models. So far, many adversarial attack and defense mechanisms have been introduced for DL models. Compromising the security and reliability of artificial intelligence (AI) systems is a major concern of government, industry and cyber-security researchers, in that insufficient protections can compromise the security and privacy of everyone in society. Finally, a defense mechanism to protect spiking neurons against adversarial attacks is introduced for the first time. In a nutshell, this dissertation presents a novel energy efficient deep spiking recurrent neural network which is inspired by delayed dynamic loops. The effectiveness of the introduced model is verified in several different applications. At the end, novel encoding and defense mechanisms are introduced which improve the robustness of the model against noise and adversarial attacks. / Doctor of Philosophy / The ultimate goal of artificial intelligence (AI) is to mimic the human brain. Artificial neural networks (ANN) are an attempt to realize that goal. However, traditional ANNs are very far from mimicking biological neurons. It is well-known that biological neurons communicate with one another through signals in the format of spikes. Therefore, artificial spiking neural networks (SNNs) have been introduced which behave more similarly to biological neurons. Moreover, SNNs are very energy efficient which makes them a suitable choice for hardware implementation of ANNs (neuromporphic computing). Despite the many benefits that are brought about by SNNs, they are still behind traditional ANNs in terms of performance. Therefore, in this dissertation, a new structure of SNNs is introduced which outperforms the traditional ANNs in three different applications. This new structure is inspired by delayed dynamic loops which exist in biological brains. The main objective of this novel structure is to capture the spatio-temporal correlation which exists in time-series while the training overhead and power consumption is reduced. Another contribution of this dissertation is to introduce novel encoding schemes for spiking neurons. It is clear that biological neurons leverage spikes, but the language that they use to communicate is not clear. Hence, the spikes require to be encoded in a certain language which is called neural spike encoding scheme. Inspired by the cognitive studies of rats, a novel encoding scheme is presented. Lastly, it is shown that the introduced encoding scheme increases the robustness of SNNs against noisy data and adversarial attacks. AI models including SNNs have shown to be vulnerable to adversarial attacks. Adversarial attacks are minor perturbations of the input data that can cause the AI model to misscalassify the data. For the first time, a defense mechanism is introduced which can protect SNNs against such attacks.

Modern robotics and automation systems require high-level reasoning capability in representing, identifying, and interpreting the three-dimensional data of the real world. Understanding the world's geometric structure by visual data is known as 3D perception. The necessity of analyzing irregular and complex 3D data has led to the development of high-dimensional frameworks for data learning. Here, we design several sparse learning-based approaches for high-dimensional data that effectively tackle multiple perception problems, including data filtering, data recovery, and data retrieval. The frameworks offer generative solutions for analyzing complex and irregular data structures without prior knowledge of data. The first part of the dissertation proposes a novel method that simultaneously filters point cloud noise and outliers as well as completing missing data by utilizing a unified framework consisting of a novel tensor data representation, an adaptive feature encoder, and a generative Bayesian network. In the next section, a novel multi-level generative chaotic Recurrent Neural Network (RNN) has been proposed using a sparse tensor structure for image restoration. In the last part of the dissertation, we discuss the detection followed by localization, where we discuss extracting features from sparse tensors for data retrieval. / Doctor of Philosophy / The development of automation systems and robotics brought the modern world unrivaled affluence and convenience. However, the current automated tasks are mainly simple repetitive motions. Tasks that require more artificial capability with advanced visual cognition are still an unsolved problem for automation. Many of the high-level cognition-based tasks require the accurate visual perception of the environment and dynamic objects from the data received from the optical sensor. The capability to represent, identify and interpret complex visual data for understanding the geometric structure of the world is 3D perception. To better tackle the existing 3D perception challenges, this dissertation proposed a set of generative learning-based frameworks on sparse tensor data for various high-dimensional robotics perception applications: underwater point cloud filtering, image restoration, deformation detection, and localization. Underwater point cloud data is relevant for many applications such as environmental monitoring or geological exploration. The data collected with sonar sensors are however subjected to different types of noise, including holes, noise measurements, and outliers. In the first chapter, we propose a generative model for point cloud data recovery using Variational Bayesian (VB) based sparse tensor factorization methods to tackle these three defects simultaneously. In the second part of the dissertation, we propose an image restoration technique to tackle missing data, which is essential for many perception applications. An efficient generative chaotic RNN framework has been introduced for recovering the sparse tensor from a single corrupted image for various types of missing data. In the last chapter, a multi-level CNN for high-dimension tensor feature extraction for underwater vehicle localization has been proposed.

A protein is a biopolymer of amino acids that encodes a particular function. Given that there are 20 amino acids possible at each site, even a short protein of 100 amino acids has $20^{100}$ possible variants, making it unrealistic to evaluate all possible sequences in sequence level space. This search space could be reduced by considering the fact that billions of years of evolution exerting a constant pressure has left us with only a small subset of protein sequences that carry out particular cellular functions. The portion of amino acid space occupied by actual proteins found in nature is therefore much smaller than that which is possible cite{kauffman1993origins}. By examining related proteins that share a conserved function and common evolutionary history (heretofore referred to as protein families), it is possible to identify common motifs that are shared. Examination of these motifs allows us to characterize protein families in greater depth and even generate new ``in silico" proteins that are not found in nature, but exhibit properties of a particular protein family. Using novel deep learning approaches and leveraging the large volume of genomic data that is now available due to high-throughput DNA sequencing, it is now possible to examine protein families in a scale and resolution that has never before been possible. By using this abundance of data to learn high dimensional representations of amino acids sequences, in this work, we show that it is possible to generate novel sequences from a particular protein family. Such a deep sequential model-based approach has great value for bioinformatics and biotechnological applications due to its rapid sampling abilities. / Master of Science / Proteins are one of the most important functional biological elements. These are composed of amino acids which link together to form different shapes which might encode a particular function. These proteins may act independently or might form ``complexes" to have a particular function. Therefore, understanding them is of utmost importance. Due to the fact that there are 20 amino acids even a protein sequence fragment of length 5 can have more than 3 million different combinations. Given, that proteins are generally 1000 amino acids long, looking at all the possibilities is next to impossible. In this work, by leveraging the ``deep learning" paradigm and the vast amount of data available, we try to model these proteins and generate new proteins belonging to a specific ``protein family." This approach has great value for bioinformatics and biotechnological applications due to its rapid sampling abilities.

With the increase in spectrum congestion, intelligent spectrum sensing systems have become more important than ever before. In the field of Radio Frequency Machine Learning (RFML), techniques like deep neural networks and reinforcement learning have been used to develop more complex spectrum sensing systems that are not reliant on expert features. Architectures like Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) have shown great promise for applications like automated modulation classification, signal detection, and specific emitter ID. Research in these areas has primarily focused on "one size fits all" networks that assume a fixed signal length in both training and inference. However, since some signals are more complex than others, due to channel conditions, transmitter/receiver effects, etc., being able to dynamically utilize just enough of the received symbols to make a reliable decision allows for more efficient decision making in applications such as electronic warfare and dynamic spectrum sharing. Additionally, the operator may want to get to the quickest possible decision. Recurrent neural networks have been shown to outperform other architectures when processing temporally correlated data, such as from wireless communication signals. However, compared to other architectures, such as CNNs, RNNs can suffer from drastically longer training and evaluation times due to their inherent sample-by-sample data processing. While traditional usage of both of these architectures typically assumes a fixed observation interval during both training and testing, the sample-by-sample processing capabilities of recurrent neural networks opens the door for "decoupling" these intervals. This is invaluable in real-world applications due to the relaxation of the typical requirement of a fixed time duration of the signals of interest. This work illustrates the benefits and considerations needed when "decoupling" these observation intervals for spectrum sensing applications. In particular, this work shows that, intuitively, recurrent neural networks can be leveraged to process less data (i.e. shorter observation intervals) for simpler inputs (less complicated signal types or channel conditions). Less intuitively, this works shows that the "decoupling" is dependent on appropriate training to avoid bias and insure generalization. / Master of Science / With the increase in spectrum congestion, intelligent spectrum sensing systems have become more important than ever before. In the field of Radio Frequency Machine Learning (RFML), techniques like deep neural networks and reinforcement learning have been used to develop more complex spectrum sensing systems that are not reliant on expert features. Architectures like Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) have shown great promise for applications like automated modulation classification, signal detection, and specific emitter ID. Research in these areas has primarily focused on "one size fits all" networks that assume a fixed signal length in both training and inference. However, since some signals are more complex than others, due to channel conditions, transmitter/receiver effects, etc., being able to dynamically utilize just enough of the received symbols to make a reliable decision allows for more efficient decision making in applications such as electronic warfare and dynamic spectrum sharing. Additionally, the operator may want to get to the quickest possible decision. Recurrent neural networks have been shown to outperform other architectures when processing temporally correlated data, such as from wireless communication signals. However, compared to other architectures, such as CNNs, RNNs can suffer from drastically longer training and evaluation times due to their inherent sample-by-sample data processing. While traditional usage of both of these architectures typically assumes a fixed observation interval during both training and testing, the sample-by-sample processing capabilities of recurrent neural networks opens the door for "decoupling" these intervals. This is invaluable in real-world applications due to the relaxation of the typical requirement of a fixed time duration of the signals of interest. This work illustrates the benefits and considerations needed when "decoupling" these observation intervals for spectrum sensing applications. In particular, this work shows that, intuitively, recurrent neural networks can be leveraged to process less data (i.e. shorter observation intervals) for simpler inputs (less complicated signal types or channel conditions). Less intuitively, this works shows that the "decoupling" is dependent on appropriate training to avoid bias and insure generalization.

A number of problems in Computer Vision, Natural Language Processing, and Machine Learning produce structured outputs in high-dimensional space, which makes searching for the global optimal solution extremely expensive. Thus, greedy algorithms, making trade-offs between precision and efficiency, are widely used. Unfortunately, they in general lack theoretical guarantees. In this thesis, we prove that greedy algorithms are effective and efficient to search for multiple top-scoring hypotheses from structured (neural) models: 1) Entropy estimation. We aim to find deterministic samples that are representative of Gibbs distribution via a greedy strategy. 2) Searching for a set of diverse and high-quality bounding boxes. We formulate this problem as the constrained maximization of a monotonic sub-modular function such that there exists a greedy algorithm having near-optimal guarantee. 3) Fill-in-the-blank. The goal is to generate missing words conditioned on context given an image. We extend Beam Search, a greedy algorithm applicable on unidirectional expansion, to bidirectional neural models when both past and future information have to be considered. We test our proposed approaches on a series of Computer Vision and Natural Language Processing benchmarks and show that they are effective and efficient. / Ph. D.