Deep reinforcement learning (RL) represents a data-driven framework for sequential decision making that has demonstrated the ability to solve challenging control tasks. This data-driven, learning-based approach offers the potential to improve operations in complex systems, but only if it can be trusted to produce reliable performance both during training and upon deployment. These requirements have hindered the adoption of deep RL in many real-world applications. In order to overcome the limitations of existing methods, this dissertation introduces reliable deep RL algorithms that deliver (i) stable training from limited data and (ii) robust, safe deployment in the presence of uncertainty. The first part of the dissertation addresses the interactive nature of deep RL, where learning requires data collection from the environment. This interactive process can be expensive, time-consuming, and dangerous in many real-world settings, which motivates the need for reliable and efficient learning. We develop deep RL algorithms that guarantee stable performance throughout training, while also directly considering data efficiency in their design. These algorithms are supported by novel policy improvement lower bounds that account for finite-sample estimation error and sample reuse. The second part of the dissertation focuses on the uncertainty present in real-world applications, which can impact the performance and safety of learned control policies. In order to reliably deploy deep RL in the presence of uncertainty, we introduce frameworks that incorporate safety constraints and provide robustness to general disturbances in the environment. Importantly, these frameworks make limited assumptions on the training process, and can be implemented in settings that require real-world interaction for training. This motivates deep RL algorithms that deliver robust, safe performance at deployment time, while only using standard data collection from a single training environment. Overall, this dissertation contributes new techniques to overcome key limitations of deep RL for real-world decision making and control. Experiments across a variety of continuous control tasks demonstrate the effectiveness of our algorithms.

No description available.

No description available.

Conventional methods for solving image restoration problems are typically built on an image degradation model and on some priors of the latent image. The model of the degraded image and the prior knowledge of the latent image are necessary because the restoration is an ill posted inverse problem. However, for some applications, such as those addressed in this thesis, the image degradation process is too complex to model precisely; in addition, mathematical priors, such as low rank and sparsity of the image signal, are often too idealistic for real world images. These difficulties limit the performance of existing image restoration algorithms, but they can be, to certain extent, overcome by the techniques of machine learning, particularly deep convolutional neural networks. Machine learning allows large sample statistics far beyond what is available in a single input image to be exploited. More importantly, the big data can be used to train deep neural networks to learn the complex non-linear mapping between the degraded and original images. This circumvents the difficulty of building an explicit realistic mathematical model when the degradation causes are complex and compounded. In this thesis, we design and implement deep convolutional neural networks (DCNN) for two challenging image restoration problems: reflection removal and joint demosaicking-deblurring. The first problem is one of blind source separation; its DCNN solution requires a large set of paired clean and mixed images for training. As these paired training images are very difficult, if not impossible, to acquire in the real world, we develop a novel technique to synthesize the required training images that satisfactorily approximate the real ones. For the joint demosaicking-deblurring problem, we propose a new multiscale DCNN architecture consisting of a cascade of subnetworks so that the underlying blind deconvolution task can be broken into smaller subproblems and solved more effectively and robustly. In both cases extensive experiments are carried out. Experimental results demonstrate clear advantages of the proposed DCNN methods over existing ones. / Thesis / Master of Applied Science (MASc)

Haze and smog are among the most common environmental factors impacting image quality and, therefore, image analysis. In this paper, I propose an end-to-end generative method for single image dehazing problem. It is based on fully convolutional network and effective network structures to recognize haze structure in input images and restore clear, haze-free ones. The proposed method is agnostic in the sense that it does not explore the atmosphere scattering model, it makes use of convolutional networks advantage in feature extraction and transfer instead. Somewhat surprisingly, it achieves superior performance relative to all existing state-of-the-art methods for image dehazing even on SOTS outdoor images, which are synthesized using the atmosphere scattering model. In order to improve its weakness in indoor hazy images and enhance the dehazed image's visual quality, a lightweight parallel network is put forward. It employs a different convolution strategy that extracts features with larger reception field to generate a complementary image. With the help of a parallel stream, the fusion of the two outputs performs better in PSNR and SSIM than other methods. / Thesis / Master of Applied Science (MASc)

We propose an end-to-end trainable Convolutional Neural Network (CNN), named GridDehazeNet, for single image dehazing. The GridDehazeNet consists of three modules: pre-processing, backbone, and post-processing. The trainable pre-processing module can generate learned inputs with better diversity and more pertinent features as compared to those derived inputs produced by hand-selected pre-processing methods. The backbone module implements a novel attention-based multi-scale estimation on a grid network, which can effectively alleviate the bottleneck issue often encountered in the conventional multi-scale approach. The post-processing module helps to reduce the artifacts in the final output. Experimental results indicate that the GridDehazeNet outperforms the state-of-the-art on both synthetic and real-world images. The proposed hazing method does not rely on the atmosphere scattering model and we provide an explanation as to why it is not necessarily beneficial to take advantage of the dimension reduction offered by the atmosphere scattering model for image dehazing, even if only the dehazing results on synthetic images are concerned. / Thesis / Master of Applied Science (MASc)

In the simple form, a communication system includes a transmitter and a receiver. In the transmitter, it transforms the one-hot vector message to produce a transmitted signal. In general, the transmitter demands restrictions on the transmitted signal. The channel is defined by the conditional probability distribution function. On receiving of the transmitted signal with noise, the receiver appears to apply the transformation to generate the estimate of one hot vector message. We can regard this simplest communication system as a specific case of autoencoder from a deep learning perspective. In our case, autoencoder used to learn the representations of the one-hot vector which are robust to the noise channel and can be recovered at the receiver with the smallest probability of error. Our task is to make some improvements on the autoencoder systems. We propose different schemes depending on the different cases. We propose a method based on optimization of softmax and introduce the L1/2 regularization in MSE loss function for SISO case and MIMO case, separately. The simulation shows that both our optimized softmax function method and L1/2 regularization loss function have a better performance than the original neural network framework. / Thesis / Master of Applied Science (MASc)

BACKGROUND: The Canadian healthcare system is subject to national standards that may be challenging to meet, given the evolution and integration of technology in healthcare in disciplines like functional neurosurgery, utilizing therapies such as deep brain stimulation (DBS), whereby implanted devices have provided benefit for patients with movement disorders. A comprehensive assessment of the need for this service to match with the delivery of DBS has not been performed. This thesis comprises a series of studies that aim to address this knowledge gap through the quadruple aim of health service research. METHODS: The first study is a systematic review and meta-analysis including economic evaluations comparing DBS for movement disorders with medical management only. The second is a mixed methods survey of Canadian stakeholders for DBS. The final study is a nationwide retrospective cohort study of DBS patients from 2019-2022 to determine factors that may influence access. RESULTS: Through analysis of 14 economic evaluations, DBS appears to be a cost-effective treatment when considered across the remaining lifespan of the patient with positive incremental net benefit for DBS with a mean difference of 40,504.81USD (95% CI 2,422.42; 78,587.19). Additionally, 220 responses from all DBS stakeholder groups revealed that costs associated with travel, waitlists, lack of specific resources, poor understanding of movement disorders and DBS indications, and referral pathways were barriers to accessing DBS. Finally, preliminary results identified 162 DBS patients. Potential factors that may increase access to DBS were indication (Parkinson’s disease), higher socioeconomic status, and race. CONCLUSIONS: While DBS is a cost-effective therapy for patients with movement disorders, the current delivery of this service needs significant improvement. This includes improved education, streamlined referral pathways, and policy change at a governmental level, with further investigation to determine regions of the country where need for DBS far exceeds current access. / Dissertation / Candidate in Philosophy / Movement disorders are progressive, debilitating neurologic conditions that severely impact the quality, speed and fluency of movement as a result of basal ganglia dysfunction. Medical therapies remain the mainstay of treatment, however high quality evidence supports the use of deep brain stimulation (DBS) to relieve these symptoms in well-selected patients. Given the upfront cost of surgery associated with DBS, and the comprehensive evaluations at tertiary care centres (including a multidisciplinary team with neurologists, neurosurgeons, neuropsychologists, psychiatrists, and electrophysiologists), this is a limited resource, particularly in overburdened publicly funded healthcare systems. There have been no previous attempts to comprehensively analyze access to DBS in Canada’s public healthcare system through investigation of need for these services, matched access, and investigation of barriers to access. This thesis comprises 5 chapters that inform this knowledge gap through the quadruple aim of health service research (patient perspective, health care provider perspective, cost, and population level data), aiming for equitable access to care in Canada. Chapter 1 is an introduction providing the rationale for conducting each of the included studies. Chapter 2 reports on an evaluation of cost, titled Economic Evaluations Comparing Deep Brain Stimulation to Best Medical Therapy for Movement Disorders: A Meta-Analysis. Chapter 3 presents an evaluation of healthcare provider and patient perspective, titled Mixed Methods Survey of Stakeholders to Identify Barriers to Accessing Deep Brain Stimulation for Movement Disorders in Canada. Chapter 4 is a retrospective cohort study providing population level data assessing patients who have received DBS in Canada, titled Canadian Access to Deep Brain Stimulation for Movement Disorders: A Nationwide Retrospective Study. Finally, Chapter 5 discusses the conclusion, limitations, and implications of the research presented in this PhD thesis.

Dissertation (Ph.D.) - Simon Fraser University, 2006. / Theses (School of Kinesiology) / Simon Fraser University. Includes bibliographical references. Also issued in digital format and available on the World Wide Web.

Thesis (M.S.)--University of Wisconsin--Madison, 1980. / Typescript. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 201-205).