About Me

I’m a CS PhD candidate (supervised by Dr. Derek Doran) at Wright State University, Dayton OH. My research interests are broadly in machine learning, with a focus on explainable and reliable deep learning, computer vision, and interaction between vision and natural languages. My academic research studies deep neural network mechanisms toward interpretability, that aid in interpreting how they process input data in a human-understandable way. I am also interested in multimodel inference and reasoning and have proposed a novel task Visual Entailment to explore challenging problems on interaction between vision and natural languages.

Contact Details

Ning Xie
xie.25@wright.edu
Wright State University
3640 Colonel Glenn Hwy
Dayton, Ohio 45435

Education

Wright State University

Ph.D. in Computer Science and Engineering GPA 4.0 Jan. 2016 - Dec. 2020(expected)

Courses Taken: Computer Vision, Algorithm Design and Analysis, Network Science, Advanced Programming Languages, Machine Learning, Smart Cities:Devices & Methods, Distributed Computing, Computer Organization, OS Internals and Design

Selected course projects:

  • Smart Cities, Devices & Methods: Augmented Convolutional Neural Network (AUG-CNN) for Autonomous Driving.
  • Network Science: Dynamic Social Network Analysis on Untrustworthy News Networks.
  • Machine Learning: Did you make the Wright Choice for your College? Using SVM to analyze college scorecard dataset public by the U.S. Department of Education.
  • Hebei University of Technology

    B.S. in Pure and Applied Mathematics Sep. 2010 - Jun. 2014

    Selected Awards:

  • National Second Prize of CUMCM (Contemporary Undergraduate Mathematics Contest Modeling), Oct. 2013
  • Learning Excellence Award in College of Science, 2012-2013
  • First-class Scholarship of the Hebei University of Technology, Nov. 2011
  • Learning model Award in College of Science, Apr. 2011
  • Experience

    Machine Learning Department, NEC Laboratories America, Inc.

    Summer Research Intern May. 2018 - Aug. 2018

  • Exploring a novel vision inference task, Visual Entailment, and build an explainable model to solve it. Work in progress. Our recent paper Visual Entailment Task for Visually-Grounded Language Learning is accepted by NeurIPS 2018 ViGIL workshop. The Visual Entailment dataset SNLI-VE is also released! [paper] [poster] [github]
  • MLCS lab, Kno.e.sis Center, Wright State University

    Graduate Research Assistant Jan. 2016 - Present

  • Currently working on explainable and reliable machine learning, a “model uncertainty” score is designed to measure the trustworthy of decisions made by the model. Work in progress.
  • Worked on Explainable Deep Learning for Computer Vision via CNN visualization techniques. One paper is accepted at NIPS 2017 Workshop on Explainable Machine Learning. [pdf] [poster]
  • Implementing deep learning systems (LSTM-RNN) to evaluate patterns in request sequence of web robot and IoT device sessions for request prediction and traffic generation. One paper on this project has been accepted by ISNN 2017. [pdf]
  • Awarded two full scholarships to attend events for women in computing: Ohio Celebration of Women in Computing at Lake Huron in Ohio, February 2017, and the CRA-W Grad Cohort Workshop in Washington, DC, April 2017
  • Poster presented at Women in STEMM Leadership Institute Symposium, Dayton, OH, Oct. 2016
  • Awarded an NSF Student Fellowship to attend 12th Reasoning Web Summer School, Aberdeen, Scotland, Sep. 2016
  • Prepared and delivered a Deep Neural Network presentation for a Wright State Seminar, Jun. 2016
  • Tianjin Tiandy Digital Technology Co., Ltd

    Algorithm Engineer Jun. 2014 - Dec. 2015

  • Intelligent Court System, Sep. 2015
    Used Deformable Parts Model (DPM) to detect people making phone calls in image data. Reduced false recognition rates by developing location and skin-color filters.
  • Floating Garbage Detection Project, May 2015
    Modified Visual Background Extractor (VIBE) to significantly improve the performance on extracting the Regions of Interest (ROI) effectively in complex environment of rivers and streams (light changing, wave interfering, etc.) to identify debris in water. Implemented an Artificial Neural Network for garbage classification.
  • Intelligent Transportation System, Oct. 2014
    Researched Pixel-Based Adaptive segmentation (PBAS) for image foreground segmentation, implementing this method for an intelligent transportation system project.
  • Publication

  • Xie, N., Lai, F., Doran, D., & Kadav, A. (2019). Visual Entailment: A Novel Task for Fine-Grained Image Understanding. arXiv preprint arXiv:1901.06706.
  • Xie, N., Lai, F., Doran, D., & Kadav, A. (2018). Visual Entailment Task for Visually-Grounded Language Learning. NeurIPS 2018 Workshop: Visually Grounded Interaction and Language (ViGIL).
  • Xie, N., Sarker, M. K., Doran, D., Hitzler, P., & Raymer, M. (2017, Dec.). Relating Input Concepts to Convolutional Neural Network Decisions. NIPS 2017 Workshop: Interpreting, Explaining and Visualizing Deep Learning, NIPS IEVDL 2017.
  • Xie, N., Brown, K., Rude, N., & Doran, D. (2017, Jun.). A Soft Computing Prefetcher to Mitigate Cache Degradation by Web Robots. In International Symposium on Neural Networks (pp. 536-546). Springer, Cham.
  • Ebrahimi, M., Sarker, M. K., Bianchi, F., Xie, N., Doran, D., & Hitzler, P. (2018) Reasoning over RDF Knowledge Bases using Deep Learning. arXiv preprint arXiv:1811.04132.
  • Sarker, M. K., Xie, N., Doran, D., Raymer, M., & Hitzler, P. Explaining Trained Neural Networks with Semantic Web Technologies: First Steps. Twelfth International Workshop on Neural-Symbolic Learning and Reasoning, NeSy (Vol. 17).
  • Check Out Some of My Recent Projects.

    Visual Entailment - A Noval Visual Reasoning Task

    In view of limitations from VQA datasets (VQA-v1.0, VQA-v2.0, CLEVR) on the pursuit of visual intelligence in the ML community, we propose a noval visual reasoning task Visual Entailment, along with a new dataset SNLI-VE. A short paper on this project has been accpeted by NeurIPS 2018 Workshop: Visually Grounded Interaction and Language (ViGIL). The full paper is available in arxiv.
    *Work performed as a NEC Labs intern.

    Explainable Deep Learning for Computer Vision

    When a person's future, health, finances, and safety is on the line, human decision makers relying on an AI system need to rationalize the recommendations the system offers. Yet present day AI is hampered by our ability to interpret or ''explain'' the rationale AI uses to arrive at a conclusion. This project investigates new approaches for reaching ''explanations'' of an AI's decision making process, specifically applied to deep learning, for computer vision tasks. A paper on this project has been accpeted by NIPS 2017 Workshop: Interpreting, Explaining and Visualizing Deep Learning, NIPS IEVDL 2017
    *Work supported by the Ohio Federal Research Network.

    Understanding the Impact of Web Robot and IoT Traffic on Web Systems

    The majority of requests seen by web servers and clouds are caused by web robots and IoT(Internet of things) agents. Prefetching web resources for caching and content management systems is a common technique to anticipate and pre-load the resources likely to be requested next for fast, low latency access. This project introduces a novel soft computing resource prefetcher, which marries a deep recurrent neural network(RNN) with a Bayesian network, to predict future requests made by web robots. Experiments with traffic logs from web servers across two universities demonstrate an improvement in precision and recall over a traditional dependency graph-based approach.
    *Work supported by the National Science Foundation under grant #1464104.

    Augmented Convolutional Neural Network for Autonomous Driving

    Autonomous driving is an active research topic these years in both academia and industry. In this work, I designed a novel end-to-end augmented convolutional neural network (AUG-CNN) to address this problem. Based on raw image pixels from the front-facing cameras equipped on the car and the car’s current status on speed and throttle, AUG-CNN autonomous driving system is able to directly make decisions to control the steering wheel and the throttle. The training dataset and testing simulator are released by Udacity Self-Driving Car Nanodegree Program. Although the AUG-CNN system is trained only on the daytime-plain-driving dataset, the performance is surprisingly good in the simulator of a nightfall-mountain-driving scenario that the model has never seen. This work is the final project for Smart Cities:Devices & Methods.

    Dynamic Social Network Analysis on Untrustworthy News Networks

    The spread of misinformation in media has become an increasingly relevant problem in society. However, automatically identifying whether a news story is an incorrect claim or factual from the content of a story is an extremely difficult machine learning problem that has yet to be solved by the research community. Our approach is to examine dynamic social networks on the discussion of news for patterns which can distinguish between disreputable claims and factual stories. Using a dynamic stochastic block model (DSBM), we demonstrate the ability to infer structural behavior such as the emergence of new connected components in social networks from temporal patterns. We also explore the potential and difficulties of using network based features to train a classifier over the truth or falseness of news. This work is the final project for Network Science



    Did you make the Wright Choice for your College?

    It is well known that U.S. university students often graduate with copious amounts of financial debt. The tuition rates for some colleges are often far above a reasonable threshold of being considered practical, which often limits many aspiring students' enrollment choices to educational institutions that seem more affordable. Even worse, students who do graduate often find themselves facing repayment obligations that far outweigh their potential income earnings. These notations collectively spark a number of questions.

  • Can we discover and explain groupings of colleges which exhibit similar characteristics?
  • Do higher tuitions colleges always provide students with higher average payment?
  • Can examining the relationship between features such as tuition costs and earnings after graduation help future students make more informed college decisions?
  • This is a final project for Machine Learning course, for more details please check the report here.

    Services

  • Program Committee of the Emoji 2019 workshop @ WWW 2019
  • Contributor of the US2TS 2019 Tutorial: On the Role of Data Semantics for Explainable AI
  • Journal reviewer for AoDSA (ARCHIVES OF DATA SCIENCE, SERIES A)
  • Conference sub-reviewer for IJCAI 2018 and ASONAM 2017