Search Constraints

Start Over
Filtering by: Creator Chen, Peter Y. S. Remove constraint Creator: Chen, Peter Y. S. Degree Field Electrical and Computer Engineering Remove constraint Degree Field: Electrical and Computer Engineering Rights Statement In Copyright Remove constraint Rights Statement: In Copyright

Search Results

  • Improving Peak Power Shaving in Data Centers using Deep Neural Networks

    Creator:
    Qi, Fang
    Abstract:
    Data centers have been charged a great amount of electric bill by the power company and demand charge can contribute up to 40% of the electric bill due to the "random" workload. This phenomenon can be avoided by using the existing Uninterrupted Power Supply (UPS) as the assistant power source...
    Resource Type:
    Capstone Project
    Full Text:
    approved: _____________________________________________________ Lizhong Chen Data centers have

  • Design techniques for low-power multi-GS/s analog-to-digital converters

    Creator:
    Jiang, Tao
    Abstract:
    Ultra-high-speed (>10GS/s), medium-resolution (5~6bit), low-power (<50mW) analog-to-digital converter can find it application in the areas of digital oscilloscopes and next-generation serial link receivers. There are several challenges to enable a successful design, however. First, the time-interleaved architecture is required in order to achieve over 10GS/s sampling rate, with the trade-off...
    Resource Type:
    Dissertation
    Full Text:
    : ______________________________________________________ Patrick Y. Chiang Ultra-high-speed (>10GS/s), medium-resolution (5~6bit), low-power

  • Leveraging transformer encoder output for effective token summary in simultaneous translation

    Creator:
    Raffel, Matthew
    Abstract:
    Simultaneous speech-to-text translation remains a difficult yet important problem for modern machine learning models whereby a text translation is generated concurrently with receiving partial speech inputs. One state-of-the-art simultaneous speech-to-text model is the augmented memory transformer whose encoder breaks a speech input into fixed-size overlapping segments composed of left, right,...
    Resource Type:
    Honors College Thesis
    Full Text:
    : Lizhong Chen Simultaneous speech-to-text translation remains a difficult yet important problem for modern

  • Machine learning for computer architecture design and optimization

    Creator:
    Penney, Drew
    Abstract:
    Machine learning applied to computer architecture has rapidly transitioned from a theoretical novelty to being a driving force behind design, control, and simulation in practically all components. These machine-learning-based methodologies are further notable for their scalability to increasingly complex design challenges, which has allowed these methodologies to surpass the prior...
    Resource Type:
    Dissertation
    Full Text:
    Chen Machine learning applied to computer architecture has rapidly transitioned from a theoretical

  • Machine Learning for Architectural Design Space Exploration and Resource Control

    Creator:
    Penney, Drew D.
    Abstract:
    Machine learning has enabled significant advancements in diverse fields, yet, with a few exceptions, has had limited impact on computer architecture. Recent work, however, has begun to explore broader application to design, optimization, and simulation. Notably, machine-learning-based strategies often surpass prior state-of-the-art analytical, heuristic, and human-expert approaches. This thesis first...
    Resource Type:
    Masters Thesis
    Full Text:
    Learning for Architectural Design Space Exploration and Resource Control Abstract approved: Lizhong Chen

  • A case for application-based linear transformers : applied towards text-to-spectrogram

    Creator:
    Agostinelli, Victor
    Abstract:
    Various natural language processing (NLP) tasks necessitate deep models that are fast, efficient, and small based on their ultimate application at the edge or elsewhere. While significant investigation has furthered the efficiency and reduced the size of these models, reducing their downstream latency without significant trade-offs remains a difficult task....
    Resource Type:
    Masters Thesis
    Full Text:
    approved: Lizhong Chen Various natural language processing (NLP) tasks necessitate deep models that are

  • Characterizing On-Chip Traffic Patterns in Throughput Processors: A Deep Learning Approach

    Creator:
    Ramamurthy, Abhishek
    Abstract:
    The machine learning and deep learning models have been very lightly explored in analyzing the behavior of On-Chip network traffic. These models have proven their potential in pattern recognition, classification etc... In this paper we analyze the spatial pattern that each workload exhibits in its life cycle during execution. We...
    Resource Type:
    Masters Thesis
    Full Text:
    . Lizhong Chen The machine learning and deep learning models have been very lightly explored in

  • A 400Mb/s 850nm FSO Link based on CMOS Integrated Receiver for Hybrid WiFi and Optical Femtocells

    Creator:
    Chen, Siyuan
    Abstract:
    The increasing demand for higher data-rates is challenging to satisfy with spectrum-deficient indoor WiFi networks. A novel hybrid WiFi Free-Space Optical (WIFO) system has been proposed to enhance the wireless capacity of indoor WiFi networks. In this thesis, an integrated optical wireless receiver is designed and integrated in 65-nm CMOS...
    Resource Type:
    Masters Thesis
    Full Text:
    supply. c©Copyright by Siyuan Chen December 04, 2018 All Rights Reserved A 400Mb/s 850nm FSO Link

  • Towards Deep Learning Model Portability for Domain-Agnostic Device Fingerprinting

    Creator:
    Gaskin, Jared
    Abstract:
    In recent years, RF (Radio Frequency) device fingerprinting using deep learning has emerged as a method of identifying devices solely by their RF transmissions. Conventional approaches to this type of device fingerprinting are not portable to different domains. That is, if a model for this purpose is trained on data...
    Resource Type:
    Masters Thesis
    Full Text:
    : Bechir Hamdaoui Lizhong Chen In recent years, RF (Radio Frequency) device fingerprinting using deep

  • TIMR : Time Interleaved Multi Rail

    Creator:
    Ruggeri, Thomas L.
    Abstract:
    This work presents a new energy saving technique for modern digital designs. We propose Time Interleaved Multi-Rail (TIMR) - a method for providing two dynamic supply rails to a circuit. This technique uses the first supply rail to mask the transition delay while changing the voltage of the second rail....
    Resource Type:
    Masters Thesis
    Full Text:
    : TIMR: Time Interleaved Multi Rail Abstract approved: Patrick Y. Chiang This work presents a new