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 26 for “"Machine Learning Hardware"”.
-
Automated design of machine learning hardware accelerators
… framework for the automated design of hardware accelerators targeting Convolutional Neural Networks (CNNs), leveraging High-Level Synthesis (HLS) and high-level model descriptions in Python (via TensorFlow or PyTorch). The proposed methodology streamlines the development of CNN dataflow …
-
Effective debug ecosystem for machine learning hardware accelerators
… have seen a dramatic increase in the use of hardware accelerators to perform machine learning computations. Designing these circuits is challenging, especially due to bugs that may only manifest after long run times, and interactions between hardware and software that are complex to …
-
Resource Allocation Optimization for Agriculture and Machine Learning Hardware Computing
… allocation strategies from agriculture to machine learning algorithms. As the world population continues growing, the food supply has become a serious problem that needs to be tread carefully. Crops play a major role in the food supply, and as crop production becomes more mechanized and …
-
FAULT MODELING, DESIGN-FOR-TEST, AND FAULT TOLERANCE FOR MACHINE LEARNING HARDWARE
… Domain-specific AI accelerators for machine-learning inferencing applications are homogeneous designs composed of thousands of identical compute cores, or processing elements (PEs), that interface with the on-chip memory (such as local and global buffers). Accelerators can be …
-
A New Approach to Machine Learning Hardware Classifier Design Based on Functional Decomposition of Multi-Valued Functions
This dissertation presents a novel design of a hardware classifier based on combining modified Ashenhurst-Curtis Decomposition and multiplexer-based synthesis. The PSUD classifier brings three new contributions: an approach to solve the column multiplicity problem, an approach to encode …
-
Ultrasound based Bi manual Gesture Detection using MVDR Beamforming and Machine Learning
… gesture detection system using beamforming and machine learning. Hardware consisting of an array of eight MEMS microphones and an ultrasonic transducer was designed and built. Software was developed to convert the hardware output into depth images of the user's gesturing hands. The depth images …
-
NEURAL INSPIRED BEHAVIORS IN ELECTRONICS
… of intelligent neural interfaces and microscopic machines, specifically focusing on applications in medical implants and autonomous microscopic robots, utilizing CMOS electronic chips. In Chapter 2, an innovative ASIC accelerator is introduced, tailored for the efficient implementation of large …
-
Accelerating Collective Communication for Distributed Machine Learning
… has emerged as a cornerstone of distributed machine learning, enabling datacenter-scale clusters of accelerators to collaboratively train or serve large language models. However, it has also become a significant performance bottleneck, impeding the efficient utilization and scalability of …
-
Compute in eDRAM using indium gallium zinc oxide transistors
With the advancement and unprecedented growth in machine learning based applications, there is a pressing need to perform energy efficient operations using application specific machine learning hardware accelerators. The major bottleneck that results in decreased performance and increased energy of …
-
FPGA Reservoir Computing Networks for Dynamic Spectrum Sensing
… beyond systems has fuelled research in merging machine learning with wireless communications to achieve cognitive radios. However, the portability and limited power supply of radio frequency devices limits engineers' ability to combine them with powerful predictive models. This hinders the …
-
Training neural networks with end-to-end optical backpropagation
… is an exciting option for the next generation of machine learning hardware that is fast, parallel and energy efficient. To create a truly all-optical neural network, it is necessary to implement both stages of deployment: inference and training. This in turn requires the ability to construct …
-
Physical design solutions for 3D ICs and their neuromorphic applications
… For the second part, we adopt the liquid-state-machine architecture, a model of recurrent spiking neural networks, to build an online machine-learning hardware platform, and study the PPA benefits of gate-level F2F and M3D ICs on the non-trivial real-world speech recognition application. This …
-
Enabling scalable & secure open-source physical design for modern VLSI
… and acceleration techniques leveraging machine learning hardware and software advancements, this the- sis lays the foundation for two open-source analytical FPGA placement tools — DREAMPlaceFPGA and aug-elfPlace, to enable scalable physical design. DREAMPlaceFPGA accelerates FPGA …
-
Compute-in-ROM : a in-memory activation-aware dot-product compute macro for low-cost edge devices
Compute-in-Memory (CiM) approaches perform multiply-and-accumulate (MAC), or dot-product, arithmetic operations required for neural network (NN) within the memory array that stores weight. Many CiM works within different memory technologies have reported superior energy efficiency compared to …
-
Hardware implementation of inference in deep neural networks
Deep learning neural network algorithms, including convolutional and recurrent networks, have risen to popularity in recent years. Along with this popularity has come a wide range of implementations that optimize the performance of these algorithms on existing hardware, including GPU architectures …
-
Efficient Modeling for DNN Hardware Resiliency Assessment
Deep neural network (DNN) hardware accelerators are critical enablers of the current resurgence in machine learning technologies. Adopting machine learning in safety-critical systems imposes additional reliability requirements on hardware design. Addressing these requirements mandates an accurate …
-
A weight-stationary systolic tensor processing unit : from architecture to physical design
The arithmetic at the heart of modern machine-learning (ML) workloads is overwhelmingly dense matrix multiplication, expressed as billions to trillions of multiply-accumulate (MAC) operations per inference. General-purpose processors expend most of their energy on instruction control and data …
-
Scalable Digital Architecture of a Liquid State Machine
Liquid State Machine (LSM) is an adaptive neural computational model with rich dynamics to process spatio-temporal inputs. These machines are extremely fast in learning because the goal-oriented training is moved to the output layer, unlike conventional recurrent neural networks. The capability to …
-
Synthesis of Smart & Intelligent Sensors
L'abstract è presente nell'allegato / the abstract is in the attachment
-
Understanding The Challenges with Using Hardware Pre-Fetchers for CPU-Based Matrix Multiply Units
Many advancements in the machine learning field have led to Graphics Processing Unit (GPU) being the de facto standard of large scale matrix computation accelerator. This work aims to answer questions about how difficult it is to use the traditional Central Processing Unit (CPU) with matrix …
Page 1 of 2