Search Constraints

Start Over
Filtering by: Creator Li, Xin Remove constraint Creator: Li, Xin Academic Affiliation Electrical Engineering and Computer Science Remove constraint Academic Affiliation: Electrical Engineering and Computer Science

Search Results

  • Don't Fool Me : Detecting Adversarial Examples in Deep Networks

    Creator:
    Li, Xin
    Abstract:
    Deep learning has greatly improved visual recognition in recent years. However, recent research has shown that there exist many adversarial examples that can negatively impact the performance of such an architecture. Different from previous perspectives that focus on improving the classifiers to detect the adversarial examples, this work focuses on...
    Resource Type:
    Masters Thesis
    Full Text:
    AN ABSTRACT OF THE THESIS OF Xin Li for the degree of Master

  • Wideband discrete-time delta-sigma analog-to-digital converters with shifted loop delays

    Creator:
    Meng, Xin
    Abstract:
    Low-distortion architecture is widely used in wideband discrete-time switched-capacitor delta-sigma ADC design. However, it suffers from the power-hungry active adder and critical timing for quantization and dynamic element matching (DEM). To solve this problem, this dissertation presents a delta-sigma modulator architecture with shifted loop delays. In this project, shifted loop...
    Resource Type:
    Dissertation
    Full Text:
    Converters With Shifted Loop Delays AN ABSTRACT OF THE THESIS OF Xin Meng for the degree of Doctor

  • Adaptive Wing Loss for Robust Face Alignment via Heatmap Regression

    Creator:
    Wang, Xinyao
    Abstract:
    Heatmap regression has became one of the mainstream approaches to localize facial landmarks. As Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) are becoming popular in solving computer vision tasks, extensive research has been done on these architectures. However, the loss function for heatmap regression is rarely studied. In...
    Resource Type:
    Masters Thesis
    Full Text:
    for Robust Face Alignment via Heatmap Regression Abstract approved: Fuxin Li Heatmap regression has

  • Toward Disentangling the Activations of the Deep Networks via Low-dimensional Embedding and Non-negative Factorization

    Creator:
    Khorram, Saeed
    Abstract:
    In this thesis, we introduce a novel Explanation Neural Network (XNN) to explain the predictions made by a deep network. The XNN works by embedding a high-dimensional activation vector of a deep network layer non-linearly into a low-dimensional explanation space while retaining faithfulness i.e., the original deep learning predictions can...
    Resource Type:
    Masters Thesis
    Full Text:
    Abstract approved: Fuxin Li In this thesis, we introduce a novel Explanation Neural Network (XNN) to

  • Methods for Detection and Recovery of Out-of-Distribution Examples

    Creator:
    Ratzlaff, Neale J.
    Abstract:
    Deep neural networks currently comprise the backbone of many applications where safety is a critical concern, for example: autonomous driving and medical diagnostics. Unfortunately these systems currently fail to detect out-of-distribution (OOD) inputs and can be prone to making dangerous errors when exposed to them. In addition, these same systems...
    Resource Type:
    Masters Thesis
    Full Text:
    Detection and Recovery of Out-of-Distribution Examples Abstract approved: Fuxin Li Deep neural networks

  • Deep Convolutional Network on Point Clouds for 3D Scene Understanding

    Creator:
    Wu, Wenxuan
    Abstract:
    As one of the most popular data types, the point cloud is widely used in various appli- cations, including computer vision, computer graphics and robotics. The capability to directly measure 3D point clouds is invaluable in those applications as depth information could remove a lot of the segmentation ambiguities in...
    Resource Type:
    Dissertation
    Full Text:
    Convolutional Network on Point Clouds for 3D Scene Understanding Abstract approved: Fuxin Li As one of the

  • Video Analysis : Techniques for Semi-Supervised Video Object Instance Segmentation and Tracking-by-Detection in the Wild

    Creator:
    Firl, Alrik K.
    Abstract:
    This thesis consists of two major components. The first part is concerned with video object instance segmentation (VOS), which is the task of assigning per-pixel labels perframe of a video sequence to indicate foreground object instance membership, given the first frame ground truth mask. VOS has myriad applications, from video...
    Resource Type:
    Masters Thesis
    Full Text:
    Abstract approved: Fuxin Li This thesis consists of two major components. The first part is concerned

  • Quantum dot plasmon coupling : Fundamental study and applications

    Creator:
    Kannegulla, Akash
    Abstract:
    The dissertation focuses on the engineering of light-matter interaction using plasmonic nanoparticles and metamaterials to achieve enhanced luminescence and based on which to improve the performance of biosensing and light-emitting technologies. We designed and fabricated a spectrum of nanostructures to exhibit particular dispersion relations capable of controlling the spontaneous emission...
    Resource Type:
    Dissertation
    Full Text:
    Abstract approved: ______________________________________________________ Li-Jing (Larry) Cheng

  • Non-enzymatic electrochemical sensors based on functionalized nanomaterials

    Creator:
    Wu, Bo
    Abstract:
    Unlike other biosensing, electrochemical biosensor has the advantages of robustness, easy miniaturization, excellent detection limits, and capability in handling small analyte volumes. The detection selectivity of the electrochemical biosensors is mainly supported by bioreceptors. However, these receptors always suffer from the performance variation caused by the environment and the thermal-...
    Resource Type:
    Dissertation
    Full Text:
    approved: _________________________________________________________________ Li-Jing (Larry) Cheng

  • Leveraging Structures of the Data in Deep Learning

    Creator:
    Li, Xingyi
    Abstract:
    The performance of deep learning frameworks could be significantly improved through considering the particular underlying structures for each dataset. In this thesis, I summarize our three work about boosting the performance of deep learning models through leveraging structures of the data. In the first work, we theoretically justify that, for...
    Resource Type:
    Dissertation
    Full Text:
    , Xinyao Wang, Neale Ratzlaff, Xin Li, Bob DeBortoli, Hung Nguyen, Weikun Han, Wenxuan Wu, Zehuan Chen