Yang Shi 石钖
Graduate student at UCIrvine
LATEST PROJECTS
Project | 01
Multi-dimensional Count Sketch
Appeared in NeurIPS 2018 workshop
When dealing with large-scale multi-dimensional data, it is crucial to design efficient compression schemes that remove redundancy while preserving relevant information. In this work, we propose multi-dimensional count sketch (MS), an unbiased estimator of the given tensor.
Project | 02
Question Type Guided Attention in Visual Question Answering
Appeared in ECCV 2018
We propose Question Type-guided Attention. It utilizes the information of question type to dynamically balance between bottom-up and top-down visual features, respectively extracted from
ResNet and Faster R-CNN networks.
[Paper] [Poster] [Codes] [Tutorial on VQA]
Project | 03
We consider the problem of robust PCA in the streaming setting with space constraints. Our result is the first to obtain finite-sample guarantees while having the weakest assumption on the sparse perturbation, namely, deterministic support, and a standard identifiability assumption on the low-rank component, namely, incoherence.
Streaming Robust PCA
Appeared in NIPS 2017 OPT Workshop
In this project, we find a intuitive and efficient way to do tensor contraction with extended BLAS kernels (Bateched BLAS). We also benchmarked with state-of-art library: BTAS.
Project | 04
Tensor Contraction using Extended BLAS Kernels
Appeared in HiPC 2016
We applied matrix robust Principle Component Analysis(PCA) to tensor cases: separate a tensor as a low rank tensor and a sparse tensor. This is done by using nonconvex optimization. We comapred our method with matrix robust PCA.
Project | 04
Project | 05
Tensor Robust PCA
Appeared in AISTATS 2016
Project | 06
Using tensor convolutional decomposition to find out usefull features for unsupervise/supervise learning.
Convolutional Tenor Decomposition in Unsupervise/Supervise Learning
Under preparation
Considering the physical water network as a graph, a CRF model is built and learned by the Structured Support Vector Machine (SSVM) . After that, we form high order CRF system by enforcing twitter leakage detection information.
Project | 07
Leak Event Identification in Water Systems Using High Order CRF
Course project