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 10 of 10 for “"Efficient Training and Inference"”.
-
On Training, Inference, and Sample Efficiencies of Language Models
… as machine translation, natural language understanding, and natural language generation. However, despite the recent developments, language models still face critical challenges. In this thesis, we investigate efficient training and inference algorithms. We also investigate the sample efficiency …
-
Federated learning for edge AI : realistic hardware-software system co-design
… enabling decentralized, privacy-preserving model training. However, real-world edge scenarios pose challenges stemming from mobility, heterogeneous resources, and limited connectivity. This dissertation addresses these gaps through a hardware-software co-design approach that aims to deliver …
-
TOWARDS CONCISE REPRESENTATION LEARNING ON DEEP NEURAL NETWORKS
… networks contain crucial information to understand the data. For example, the image representations from Convolutional Neural Networks (CNNs) can express the low-level and high-level features from the images; the video representations from deep neural networks convey both spatial and temporal …
-
Optimal transport based simulation methods for deep probabilistic models
… for high-dimensional, multi-modal data synthesis and density estimation tasks. By combining abstract probabilistic formulations with the expressivity and scalability of neural networks, deep probabilistic models have become a fundamental component of the machine learning toolbox. Such models still …
-
Early-bird training and hierarchical hardware-aware model compression towards green and ubiquitous artificial intelligence
… such as image perception, augmented and virtual reality (AR/VR), and AI-generated content (AIGC). However, critical research gaps remain between the powerful yet large-scale AI models and the computational constraints of both edge and cloud platforms, including (1) the efficiency gap, …
-
Efficient deep learning methodologies
… contributions to both academic research and industry, leading to state-of-the-art performance in a wide range of applications. However, DL models increasingly require more powerful hardware, raising concerns regarding their sustainability. To address the demanding nature of DL and meet …
-
Efficient machine learning software stack from algorithms to compilation
… enables the extraction of knowledge from data and decision-making without explicit programming, achieving great success and revolutionizing many fields. These successes can be attributed to the continuous advancements in machine learning software and hardware, which have expanded the boundaries …
-
Compression and communication of data for next-generation connected systems
… in an increasingly distributed society that demands constant connectivity, enabling people to work, communicate, and collaborate across geographic boundaries. Sustaining this connectivity rests on two key pillars. The first is the reliable and efficient transmission of data over communication …
-
Non-parametric Bayesian approaches to deep neural networks
… comprises the construction of more flexible and efficient models, aspiring to increase the modeling capacity while addressing the complexity issues of the conventionally employed approaches. The thesis initially focuses on the batch scenario, where we consider Deep Neural architectures. …
-
Continual Learning of Object Classification in the Real World
… object classification task but only when all the training data of classes to be learned are available at the same time. However, real-world data continually evolve through time, resulting in ever-changing learning configurations, e.g., new classes are added continuously. When a deep learning model …