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 20 of 24 for “"quantization-aware training"”.
-
Treinamento consciente de quantização vetorial evolutiva: um método adaptativo para compactar modelos de aprendizado de máquina
… techniques, predominantly applied after training, prove inadequate in this context, as they may degrade model performance or rely on opaque execution pipelines, hindering the analysis and verification of system behavior. In response to these challenges, this thesis proposes Evolving …
-
An Exploration of U-Net-Based Cloud Segmentation in the Bayer Domain for Satellite Imagery
… model compression through structural pruning and quantization. Structural pruning is implemented by progressively reducing the network depth to assess the trade-off between model complexity and segmentation accuracy. For the quantization, both post-training quantization and quantization-aware …
-
Efficient deep learning methodologies
… advanced methodologies for both efficient training and inference, as well as for designing specialized hardware targeting increased speed and reduced energy consumption. Motivated by the fact that the advent of DL became feasible only when the appropriate hardware was designed to support …
-
Generalized Machine Learning Quantization Implementation for High Level Synthesis Targeting FPGAs
… acceleration options. FINN utilizes Brevitas, a quantization aware training library. Using Brevitas, I trained a particle tracking network and demonstrated equivalent accuracy at lower bit precision than post training quantization. As a cross organizational project, the hls4ml and FINN teams …
-
Hardware-Algorithm Co-Design for Energy-Efficient and Low-Latency Domain-Specific Machine Learning Systems
… for efficient feature extraction and employ Quantization-Aware Training (QAT) with low-Hamming-weight binary representations, significantly reducing inference energy costs. At the neural network level, we design ChirpNet, a radar-specific architecture that sequentially processes Frequency …
-
Optimization of Deep Neural Networks for Efficient Resource Utilization
… edge devices. This thesis explores quantization as a model compression technique to improve inference efficiency while minimizing accuracy loss. Post-Training Quantization (PTQ) and Quantization-Aware Training (QAT) were applied to MobileNetV2 and ResNet50 for classification on the …
-
Convolutional Neural Networks on the Edge
… range of experiments, we have learned that post quantization and quantization aware training models can score equally well on the UrbanSound8k dataset compared to floating point models. Acoustic event detection models can characterize an acoustic environment where the scene classification score …
-
Trainable Fixed-Point Quantization for Deep Learning Acceleration on FPGAs
This thesis introduces a novel fixed-point quantization aware training tool, QFX, designed to bridge the gap between trained and deployed deep learning models on resource-constrained devices, such as embedded FPGAs. Conventional quantization techniques have primarily focused on quantizing only the …
-
Optimization of battery AI models for edge deployment: A systematic study of model compression techniques and hardware-aware efficiency
… resources. This thesis evaluates a hardware-aware workflow for transferring, compressing, and preparing a long short-term memory (LSTM)-based SoH estimation model for embedded execution. The baseline model and its original training were provided by Cognivity AI; model development is therefore …
-
Resource optimization methods for Machine Learning model execution on resource constrained embedded systems
… study of key model compression techniques -quantization, pruning and factorization- and their impact on inference performance and energy consumption on modern embedded platforms. Theoretical discussions in Chapters 1 and 2 explore the evolution of embedded intelligence, the logic of neural …
-
Mitigating computational bottlenecks in deep neural networks through reduced-precision hardware and software
… the accuracy loss characteristic of aggressive quantization, we offered compensatory training methods, including device-aware, bit-slicing-aware, and PS-quantization-aware training for backpropagation across stochastic PSs, as well as inhomogeneous sampling rates for column-wise stochastic …
-
Energy-Aware TinyML for Ambient-Powered, Hardware-Constrained IoT Nodes
… models were further optimized using post-training quantization (PTQ), quantization-aware training (QAT), and a two-stage magnitude-based pruning method. Different optimization orders were compared with respect to model size, reconstruction quality, classifier accuracy, and estimated …
-
Data-Driven Plant Models for Hydraulic Excavators: Development of Compact Machine Learning Architectures across Simulated and Physical Platforms
… model compression via knowledge distillation and quantization-aware training was applied. The resulting compressed architecture achieved a model size of 24K parameters while only increasing prediction errors by ≈ 30%. This work demonstrates that hardware-aware deep learning models can serve as a …
-
Automated data analysis for bridge visual inspection using deep learning techniques
… was developed using the partial quantization-aware training technique. Comparison experiments indicated that the quantized model expedited the detection speed by 83.5% and reduced energy consumption by 79.7% while still maintaining a comparable level of detection accuracy. The …
-
On the Design of Efficient Deep Learning Methods for Human Activity Recognition in Resource Constrained Devices
… from various modalities of data. However, the training and inference of CNNs are often resource-intensive. Recent research developments are focused on bringing the effectiveness of CNNs in resource constrained edge devices through Tiny Machine Learning (TinyML). TinyML aims to optimize these …
-
Energy-efficient Neuromorphic Computing for Resource-constrained Internet of Things Devices
… four essential methods. The first method is the quantization of neural networks through knowledge distillation. This work introduces a quantization technique that effectively reduces the computational and storage resource requirements of a model while minimizing the loss of accuracy. To further …
-
IMPROVING MODEL AND DATA EFFICIENCY FOR DEEP LEARNING
… such achievements are predominantly derived from training huge models (i.e., billions of parameters) on numerous labeled examples, which requires considerable computation resources and expensive data collection costs. Various studies have strived to enhance efficiency in these domains. In terms of …
-
Application of Large Language Models in Code Clone: Benchmarking, Evaluation and Detection
… model using knowledge distillation (KD) and INT8 quantization-aware training (QAT). Our approach compressed LLM model, significantly reducing the number of attention heads, hidden dimensions, and layers while maintaining high performance. To compare with the existing LLM-tools which were trained …
-
A Deep Learning-Based Approach for Detection and Tracking of IoT Botnet Activities
… of IoT traffic caused by concept drift. Finally, quantization is deployed for model compression to allow deployment on the resource constrained IoT edge devices. Collectively, this thesis makes the following key contributions toward enhancing detection efficiency in the highly dynamic and evolving …
-
Machine Learning for Perception and Localization: Efficient and Invariant Methods
… In Paper IV, we therefore explore floating point quantization of neural networks in the context of federated learning and propose a new method that allows training to be performed on low-precision hardware. More specifically, we propose a method for quantization aware training and server-to-device …
Page 1 of 2