In the rapidly evolving field of computer vision, propelled by advancements in deeplearning, the integration of hardware-software co-design has become crucial to overcome the limitations of traditional imaging systems. This dissertation explores the integration of hardware-software co-design in computational imaging, particularly in light transport acquisition and Non-Line-of-Sight (NLOS) imaging. By leveraging projector-camera systems and computational techniques, this thesis address critical challenges in imaging complex environments, such as adverse weather conditions, low-light scenarios, and the imaging of reflective or transparent objects. The first contribution in this thesis is the theory, design, and implementation of a slope disparity gating system, which is a vertically aligned configuration of a synchronized raster scanning projector and rolling-shutter camera, facilitating selective imaging through disparity-based triangulation. This system introduces a novel, hardware-oriented approach to selective imaging, circumventing the limitations of post-capture processing. The second contribution of this thesis is the realization of two innovative approaches for spotlight optimization to improve localization and tracking for NLOS imaging. The first approach utilizes radiosity-based optimization to improve 3D localization and object identification for small-scale laboratory settings. The second approach introduces a learningbased illumination network along with a differentiable renderer and NLOS estimation network to optimize human 2D localization and activity recognition. This approach is validated on a large, room-scale scene with complex line-of-sight geometries and occluders. The third contribution of this thesis is an attention-based neural network for passive NLOS settings where there is no controllable illumination. The thesis demonstrates realtime, dynamic NLOS human tracking where the camera is moving on a mobile robotic platform. In addition, this thesis contains an appendix featuring temporally consistent relighting for portrait videos with applications in computer graphics and vision.

This thesis examines the critical relationship between data, complex models, and other methods to measure and analyze them. As models grow larger and more intricate, they require more data, making it vital to use that data effectively. The document starts with a deep dive into nonconvex functions, a fundamental element of modern complex systems, identifying key conditions that ensure these systems can be analyzed efficiently—a crucial consideration in an era of vast amounts of variables. Loss functions, traditionally seen as mere optimization tools, are analyzed and recast as measures of how accurately a model reflects reality. This redefined perspective permits the refinement of data-sourcing strategies for a better data economy. The aim of the investigation is the model itself, which is used to understand and harness the underlying patterns of complex systems. By incorporating structure both implicitly (through periodic patterns) and explicitly (using graphs), the model's ability to make sense of the data is enhanced. Moreover, online learning principles are applied to a crucial practical scenario: robotic resource monitoring. The results established in this thesis, backed by simulations and theoretical proofs, highlight the advantages of online learning methods over traditional ones commonly used in robotics. In sum, this thesis presents an integrated approach to measuring complex systems, providing new insights and methods that push forward the capabilities of machine learning.

With the significant advancements of wireless communication systems that aim to meet exponentially increasing data rate demands, two promising concepts have appeared: (i) Cell-free massive MIMO, which entails the joint transmission and processing of the signals allowing the removal of classical cell boundaries, and (ii) integrated sensing and communication (ISAC), unifying communication and sensing in a single framework. This dissertation aims to take steps toward overcoming the key challenges in each concept and eventually merge them for efficient future communication and sensing networks.Cell-free massive MIMO is a distributed MIMO concept that eliminates classical cell boundaries and provides a robust performance. A significant challenge in realizing the cell-free massive MIMO in practice is its deployment complexity. In particular, connecting its many distributed access points with the central processing unit through wired fronthaul is an expensive and time-consuming approach. To eliminate this problem and enhance scalability, in this dissertation, a cell-free massive MIMO architecture adopting a wireless fronthaul is proposed, and the optimization of achievable rates for the end-to-end system is carried out. The evaluation has shown the strong potential of employing wireless fronthaul in cell-free massive MIMO systems. ISAC merges radar and communication systems, allowing effective sharing of resources, including bandwidth and hardware. The ISAC framework also enables sensing to aid communications, which shows a significant potential in mobile communication applications. Specifically, radar sensing data can address challenges like beamforming overhead and blockages associated with higher frequency, large antenna arrays, and narrow beams. To that end, this dissertation develops radar-aided beamforming and blockage prediction approaches using low-cost radar devices and evaluates them in real-world systems to verify their potential. At the intersection of these two paradigms, the integration of sensing into cell-free massive MIMO systems emerges as an intriguing prospect for future technologies. This integration, however, presents the challenge of considering both sensing and communication objectives within a distributed system. With the motivation of overcoming this challenge, this dissertation investigates diverse beamforming and power allocation solutions. Comprehensive evaluations have shown that the incorporation of sensing objectives into joint beamforming designs offers substantial capabilities for next-generation wireless communication and sensing systems.

Event identification is increasingly recognized as crucial for enhancing the reliability, security, and stability of the electric power system. With the growing deployment of Phasor Measurement Units (PMUs) and advancements in data science, there are promising opportunities to explore data-driven event identification via machine learning classification techniques. This dissertation explores the potential of data-driven event identification through machine learning classification techniques. In the first part of this dissertation, using measurements from multiple PMUs, I propose to identify events by extracting features based on modal dynamics. I combine such traditional physics-based feature extraction methods with machine learning to distinguish different event types.Using the obtained set of features, I investigate the performance of two well-known classification models, namely, logistic regression (LR) and support vector machines (SVM) to identify generation loss and line trip events in two datasets. The first dataset is obtained from simulated events in the Texas 2000-bus synthetic grid. The second is a proprietary dataset with labeled events obtained from a large utility in the USA. My results indicate that the proposed framework is promising for identifying the two types of events in the supervised setting. In the second part of the dissertation, I use semi-supervised learning techniques, which make use of both labeled and unlabeled samples.I evaluate three categories of classical semi-supervised approaches: (i) self-training, (ii) transductive support vector machines (TSVM), and (iii) graph-based label spreading (LS) method. In particular, I focus on the identification of four event classes i.e., load loss, generation loss, line trip, and bus fault. I have developed and publicly shared a comprehensive Event Identification package which consists of three aspects: data generation, feature extraction, and event identification with limited labels using semi-supervised methodologies. Using this package, I generate eventful PMU data for the South Carolina 500-Bus synthetic network. My evaluation confirms that the integration of additional unlabeled samples and the utilization of LS for pseudo labeling surpasses the outcomes achieved by the self-training and TSVM approaches. Moreover, the LS algorithm consistently enhances the performance of all classifiers more robustly.

With the proliferation of mobile computing and Internet-of-Things (IoT), billions of mobile and IoT devices are connected to the Internet, generating zillions of Bytes of data at the network edge. Driving by this trend, there is an urgent need to push the artificial intelligence (AI) frontiers to the network edge to unleash the potential of the edge big data fully. This dissertation aims to comprehensively study collaborative learning and optimization algorithms to build a foundation of edge intelligence. Under this common theme, this dissertation is broadly organized into three parts. The first part of this study focuses on model learning with limited data and limited computing capability at the network edge. A global model initialization is first obtained by running federated learning (FL) across many edge devices, based on which a semi-supervised algorithm is devised for an edge device to carry out quick adaptation, aiming to address the insufficiency of labeled data and to learn a personalized model efficiently. In the second part of this study, collaborative learning between the edge and the cloud is studied to achieve real-time edge intelligence. More specifically, a distributionally robust optimization (DRO) approach is proposed to enable the synergy between local data processing and cloud knowledge transfer. Two attractive uncertainty models are investigated corresponding to the cloud knowledge transfer: the distribution uncertainty set based on the cloud data distribution and the prior distribution of the edge model conditioned on the cloud model. Collaborative learning algorithms are developed along this line. The final part focuses on developing an offline model-based safe Inverse Reinforcement Learning (IRL) algorithm for connected Autonomous Vehicles (AVs). A reward penalty is introduced to penalize unsafe states, and a risk-measure-based approach is proposed to mitigate the model uncertainty introduced by offline training. The experimental results demonstrate the improvement of the proposed algorithm over the existing baselines in terms of cumulative rewards.

In this work, the author analyzes quantitative and structural aspects of Bayesian inference using Markov kernels, Wasserstein metrics, and Kantorovich monads. In particular, the author shows the following main results: first, that Markov kernels can be viewed as Borel measurable maps with values in a Wasserstein space; second, that the Disintegration Theorem can be interpreted as a literal equality of integrals using an original theory of integration for Markov kernels; third, that the Kantorovich monad can be defined for Wasserstein metrics of any order; and finally, that, under certain assumptions, a generalized Bayes’s Law for Markov kernels provably leads to convergence of the expected posterior distribution in the Wasserstein metric. These contributions provide a basis for studying further convergence, approximation, and stability properties of Bayesian inverse maps and inference processes using a unified theoretical framework that bridges between statistical inference, machine learning, and probabilistic programming semantics.

This dissertation explores the use of artificial intelligence and machine learningtechniques for the development of controllers for fully-powered robotic prosthetics. The aim of the research is to enable prosthetics to predict future states and control biomechanical properties in both linear and nonlinear fashions, with a particular focus on ergonomics. The research is motivated by the need to provide amputees with prosthetic devices that not only replicate the functionality of the missing limb, but also offer a high level of comfort and usability. Traditional prosthetic devices lack the sophistication to adjust to a user’s movement patterns and can cause discomfort and pain over time. The proposed solution involves the development of machine learning-based controllers that can learn from user movements and adjust the prosthetic device’s movements accordingly. The research involves a combination of simulation and real-world testing to evaluate the effectiveness of the proposed approach. The simulation involves the creation of a model of the prosthetic device and the use of machine learning algorithms to train controllers that predict future states and control biomechanical properties. The real- world testing involves the use of human subjects wearing the prosthetic device to evaluate its performance and usability. The research focuses on two main areas: the prediction of future states and the control of biomechanical properties. The prediction of future states involves the development of machine learning algorithms that can analyze a user’s movements and predict the next movements with a high degree of accuracy. The control of biomechanical properties involves the development of algorithms that can adjust the prosthetic device’s movements to ensure maximum comfort and usability for the user. The results of the research show that the use of artificial intelligence and machine learning techniques can significantly improve the performance and usability of pros- thetic devices. The machine learning-based controllers developed in this research are capable of predicting future states and adjusting the prosthetic device’s movements in real-time, leading to a significant improvement in ergonomics and usability. Overall, this dissertation provides a comprehensive analysis of the use of artificial intelligence and machine learning techniques for the development of controllers for fully-powered robotic prosthetics.

Quantum computing is becoming more accessible through modern noisy intermediate scale quantum (NISQ) devices. These devices require substantial error correction and scaling before they become capable of fulfilling many of the promises that quantum computing algorithms make. This work investigates the current state of NISQ devices by implementing multiple classical computing scenarios with a quantum analog to observe how current quantum technology can be leveraged to achieve different tasks. First, quantum homomorphic encryption (QHE) is applied to the quantum teleportation protocol to show that this form of algorithm security is possible to implement with modern quantum computing simulators. QHE is capable of completely obscuring a teleported state with a liner increase in the number of qubit gates O(n). Additionally, the circuit depth increases minimally by only a constant factor O(c) when using only stabilizer circuits. Quantum machine learning (QML) is another potential application of NISQ technology that can be used to modify classical AI. QML is investigated using quantum hybrid neural networks for the classification of spoken commands on live audio data. Additionally, an edge computing scenario is examined to profile the interactions between a quantum simulator acting as a cloud server and an embedded processor board at the network edge. It is not practical to embed NISQ processors at a network edge, so this paradigm is important to study for practical quantum computing systems. The quantum hybrid neural network (QNN) learned to classify audio with equivalent accuracy (~94%) to a classical recurrent neural network. Introducing quantum simulation slows the systems responsiveness because it takes significantly longer to process quantum simulations than a classical neural network. This work shows that it is viable to implement classical computing techniques with quantum algorithms, but that current NISQ processing is sub-optimal when compared to classical methods.

Computed tomography (CT) and synthetic aperture sonar (SAS) are tomographic imaging techniques that are fundamental for applications within medical and remote sensing. Despite their successes, a number of factors constrain their image quality. For example, a time-varying scene during measurement acquisition yields image artifacts. Additionally, factors such as bandlimited or sparse measurements limit image resolution. This thesis presents novel algorithms and techniques to account for these factors during image formation and outperform traditional reconstruction methods. In particular, this thesis formulates analysis-by-synthesis optimizations that leverage neural fields to predict the scene and differentiable physics models that incorporate prior knowledge about image formation. The specific contributions include: (1) a method for reconstructing CT measurements from time-varying (non-stationary) scenes; (2) a method for deconvolving SAS images, which benefits image quality; (3) a method that couples neural fields and a differentiable acoustic model for 3D SAS reconstructions.

Quantum computing holds the potential to revolutionize various industries by solving problems that classical computers cannot solve efficiently. However, building quantum computers is still in its infancy, and simulators are currently the best available option to explore the potential of quantum computing. Therefore, developing comprehensive benchmarking suites for quantum computing simulators is essential to evaluate their performance and guide the development of future quantum algorithms and hardware. This study presents a systematic evaluation of quantum computing simulators’ performance using a benchmarking suite. The benchmarking suite is designed to meet the industry-standard performance benchmarks established by the Defense Advanced Research Projects Agency (DARPA) and includes standardized test data and comparison metrics that encompass a wide range of applications, deep neural network models, and optimization techniques. The thesis is divided into two parts to cover basic quantum algorithms and variational quantum algorithms for practical machine-learning tasks. In the first part, the run time and memory performance of quantum computing simulators are analyzed using basic quantum algorithms. The performance is evaluated using standardized test data and comparison metrics that cover fundamental quantum algorithms, including Quantum Fourier Transform (QFT), Inverse Quantum Fourier Transform (IQFT), Quantum Adder, and Variational Quantum Eigensolver (VQE). The analysis provides valuable insights into the simulators’ strengths and weaknesses and highlights the need for further development to enhance their performance. In the second part, benchmarks are developed using variational quantum algorithms for practical machine learning tasks such as image classification, natural language processing, and recommendation. The benchmarks address several unique challenges posed by benchmarking quantum machine learning (QML), including the effect of optimizations on time-to-solution, the stochastic nature of training, the inclusion of hybrid quantum-classical layers, and the diversity of software and hardware systems. The findings offer valuable insights into the simulators’ ability to solve practical machine-learning tasks and pinpoint areas for future research and enhancement. In conclusion, this study provides a rigorous evaluation of quantum computing simulators’ performance using a benchmarking suite that meets industry-standard performance benchmarks.