Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 13 of 13 for “"On-device inference"”.
-
Algorithms and Theory for Saddle Point Problems with Relaxed Structural Conditions
Saddle point, also known as min--max, optimization considers problems in which one variable is minimized while another is maximized, typically through an objective that couples the two variables. These problems arise naturally in areas including game theory, robust optimization, and adversarial …
-
Adaptive multimodal smart home control on a raspberry pi 5 using hand gestures, voice cues, and user feedback
<p>A real time multimodal smart home control system deployed on a Raspberry Pi 5 is presented. The system combines hand gestures, short voice cues, and proximity aware interaction to execute household commands such as light brightness control, fan speed adjustment, and stop or kill switch actions. …
-
Domain-Specific Design on FPGA and NPU for scientific computing and On-Device LLM Inference
… (HPC) is changing rapidly as scientific simulations and large language model (LLM) workloads push the need for higher performance under tight power and memory constraints. Conventional platforms such as CPUs and GPUs accelerate computation through instruction-driven parallelism, relying on …
-
AI-Driven Penetration Testing for Arm Systems: A Comprehensive Framework With Experimental Validation
The convergence of artificial intelligence and cybersecurity presents new opportunities for automated penetration testing capable of discovering, prioritizing, and remediating vulnerabilities at machine speed. However, deployment on resource-constrained ARM platforms remains unexplored despite …
-
MiniVO: A Lightweight Visual Odometry Model for Resource-Constrained Systems
In both biology and robotics, vision is fundamental to navigation and autonomy. Just as animals rely on visual cues to move through their environments, avoid obstacles, and reach goals, autonomous systems depend on visual processing to perceive and interact with the world. Visual Odometry (VO) …
-
Hardware-Friendly Model Compression techniques for Deep Learning Accelerators
… of the proposed research is to introduce solutions to make energy-efficient \gls{dnn} accelerators to be deployable on edge devices through developing hardware-aware \gls{dnn} compression methods. The rising popularity of intelligent mobile devices and the computational cost of deep …
-
Efficient deep neural network inference for embedded systems : A mixture of experts approach
Deep neural networks (DNNs) have become one of the dominant machine learning approaches in recent years for many application domains. Unfortunately, DNNs are not well suited to addressing the challenges of embedded systems, where on-device inference on battery-powered, resource-constrained devices …
-
Nova: Privacy-Preserving Goal-Oriented Reasoning in Smart Homes with On-Device Vision-Language Models
… under-specified user commands with the introduction of Large Language Models (LLMs). However, goal-oriented commands such as "Get this space ready for a party" expose a fundamental limitation in current systems. Accomplishing such goals requires coordinating three distinct categories of tasks: …
-
Towards Securing Quantized Deep Learning
… facilitated a wide array of practical applications, spanning from image recognition, autonomous driving, to speech processing. The significance of DL is further underscored by recent advancements in large language models such as ChatGPT, Bard, LLaMA, and in image-generative Artificial …
-
Leveraging machine learning and computer vision for advanced UAV communications
… rapidly developing field of wireless communication, there is a growing demand for technologies that can provide flexible deployment, extended coverage, and enhanced performance in next-generation networks. Traditional networks often struggle with high mobility and environmental blockages, …
-
Energy-efficient systems for information transfer and processing
… data deluge of the big data era thanks to the exponential increase in computing power. Current ML systems adopt either centralized cloud computing or distributed edge computing. In both, the challenge of energy efficiency has been drawing increased attention. In cloud computing, data transfer due …
-
Mobilní aplikace pro detekci a identifikaci vizuálně podobných objektů v obraze
… obrazu. Niekoľko ľahkých embeddingových backbone architektúr a stratových funkcií bolo vyhodnotených a doladených na dátovej sade zameranej na rastliny. Výsledné modely boli následne optimalizované pomocou kvantizácie a prerezávania a nasadené prostredníctvom vhodných frameworkov pre …
-
Efficient deep learning in mobile and embedded computing environments
… as natural language processing, computer vision, and autonomous decision-making. However, the ever-increasing complexity of modern models entails substantial computational demands, rendering the use of powerful cloud infrastructures essential. This dependence on centralized computing …