The document summarizes the background and qualifications of Zejia Zheng. It outlines his expertise in areas such as artificial intelligence, machine learning, and neural networks. It also lists his academic and professional experience, including a Ph.D from Michigan State University in computer science and internships at Zoox and Samsung Research America developing deep learning and image classification systems. His research focuses on areas like reinforcement learning, mobile device navigation, and computational neuroscience modeling.
A description of software as infrastructure at NSF, and how Apache projects may be similar. What lessons can be shared from one organization to the other? How does science software compare with more general software?
EEG Based BCI Applications with Deep LearningRiddhi Jain
Summarised a Survey Paper describing EEG Based BCI Applications and Sensing Technologies and their Computational Intelligence Approach published on Jan 28, 2020
Cognitive Computing by Professor Gordon Pipadiannepatricia
Professor Dr. Gordon Pipa, University of Osnabrueck, Germany is making this presentation for the Cognitive Systems Institute Speaker Series on May 26, 2016.
Reproducibility in human cognitive neuroimaging: a community-driven data sha...Nolan Nichols
Access to primary data and the provenance of derived data are increasingly recognized as an essential aspect of reproducibility in biomedical research. While productive data sharing has become the norm in some biomedical communities, human brain imaging has lagged in open data and descriptions of provenance. The overarching goal of my dissertation was to identify barriers to neuroimaging data sharing and to develop a fundamentally new, granular data exchange standard that incorporates provenance as a primitive to document cognitive neuroimaging workflow.
For my dissertation research, I led the development of the Neuroimaging Data Model (NIDM), an extension to the W3C PROV standard for the domain of human brain imaging. NIDM provides a language to communicate provenance by representing primary data, computational workflow, and derived data as bundles of linked Agents, Activities, and Entities. Similar to the way a sentence conveys a standalone thought, a bundle contains provenance statements that parsimoniously express the way a given piece of data was produced. To demonstrate a system that implements NIDM, I developed a modern, semantic Web application platform that provides neuroimaging workflow as a service and captures provenance statements as NIDM bundles. The course of this work necessitated interaction with an international community, which adopted and extended central elements of this work into prevailing brain imaging software. My dissertation contributes neuroinformatics standards to advance the current state of computational infrastructure available to the cognitive neuroimaging community.
Presented at OECD Workshop on Systematic Reviews in the Scope of the Endocrine Disrupter Testing and Assessment (EDTA) Conceptual Framework Level 1 in Paris, France
Artificial Neural Network Seminar - Google BrainRawan Al-Omari
it's our seminar in artificial neural network course, at F.I.T.E, AI Dept.
it's about Google Brain project, and who they using neural network in building it .
actually it's a very interesting project they work on it .
for more information about this project :
http://nyti.ms/T5E71e
A description of software as infrastructure at NSF, and how Apache projects may be similar. What lessons can be shared from one organization to the other? How does science software compare with more general software?
EEG Based BCI Applications with Deep LearningRiddhi Jain
Summarised a Survey Paper describing EEG Based BCI Applications and Sensing Technologies and their Computational Intelligence Approach published on Jan 28, 2020
Cognitive Computing by Professor Gordon Pipadiannepatricia
Professor Dr. Gordon Pipa, University of Osnabrueck, Germany is making this presentation for the Cognitive Systems Institute Speaker Series on May 26, 2016.
Reproducibility in human cognitive neuroimaging: a community-driven data sha...Nolan Nichols
Access to primary data and the provenance of derived data are increasingly recognized as an essential aspect of reproducibility in biomedical research. While productive data sharing has become the norm in some biomedical communities, human brain imaging has lagged in open data and descriptions of provenance. The overarching goal of my dissertation was to identify barriers to neuroimaging data sharing and to develop a fundamentally new, granular data exchange standard that incorporates provenance as a primitive to document cognitive neuroimaging workflow.
For my dissertation research, I led the development of the Neuroimaging Data Model (NIDM), an extension to the W3C PROV standard for the domain of human brain imaging. NIDM provides a language to communicate provenance by representing primary data, computational workflow, and derived data as bundles of linked Agents, Activities, and Entities. Similar to the way a sentence conveys a standalone thought, a bundle contains provenance statements that parsimoniously express the way a given piece of data was produced. To demonstrate a system that implements NIDM, I developed a modern, semantic Web application platform that provides neuroimaging workflow as a service and captures provenance statements as NIDM bundles. The course of this work necessitated interaction with an international community, which adopted and extended central elements of this work into prevailing brain imaging software. My dissertation contributes neuroinformatics standards to advance the current state of computational infrastructure available to the cognitive neuroimaging community.
Presented at OECD Workshop on Systematic Reviews in the Scope of the Endocrine Disrupter Testing and Assessment (EDTA) Conceptual Framework Level 1 in Paris, France
Artificial Neural Network Seminar - Google BrainRawan Al-Omari
it's our seminar in artificial neural network course, at F.I.T.E, AI Dept.
it's about Google Brain project, and who they using neural network in building it .
actually it's a very interesting project they work on it .
for more information about this project :
http://nyti.ms/T5E71e
Comment un tableau de bord smart City peut être au service de la performance publique ? Comment maintenir la participation du citoyen et éviter une approche utilitariste ? Comment créer une plateforme collaborative permettant aux différents acteurs de s'investir dans des missions de service public ? Telles sont les questions qui se jouent. L'exploration de ce sujet repose sur une idée simple, selon laquelle les élus, les services de la ville, les entreprises et les citoyens doivent construire ensemble une ville plus intelligente.
NumPyCNNAndroid: A Library for Straightforward Implementation of Convolutiona...Ahmed Gad
The presentation of my paper titled "#NumPyCNNAndroid: A Library for Straightforward Implementation of #ConvolutionalNeuralNetworks for #Android Devices" at the second International Conference of Innovative Trends in #ComputerEngineering (ITCE 2019).
The paper proposes a library for implementing convolutional neural networks (CNNs) in order to run on Android devices. The process of running the CNN on the mobile devices is straightforward and does not require an in-between step for model conversion as it uses #Kivy cross-platform library.
The CNN layers are implemented in #NumPy. You can find their implementation in my #GitHub project at this link: https://github.com/ahmedfgad/NumPyCNN
The library is also open source available here: https://github.com/ahmedfgad/NumPyCNNAndroid
There are 2 modes of operation for this work. The first one is training the CNN on the mobile device but it is very time-consuming at least in the current version. The second and preferred way is to train the CNN in a desktop computer and then use it on the mobile device.
1. ARTIFICIAL INTELLIGENCE | MACHINE LEARNING| DEEP LEARNING NEURAL NETWORKS |
MATHEMATICAL MODELING | PROGRAMMING SKILLS | COMMUNICATIVE SKILLS
KEY ACCOMPLISHMENTS
• Recognized expertise in Artificial Intelligence, Machine Learning, Algorithms Design and Analysis, and Mathematical
Modeling.
• Experienced in deep learning networks, Stacked Auto-encoders, Convolutional Neural Networks and hierarchical visual
cortex models for image classification and object recognition.
• Expert in state based reinforcement learning concepts, mechanisms, and techniques, and on-line learning neural network
based neuromodulatory system design for reinforcement learning.
• Skilled in Android application programming, mobile application user interface design, mobile device based image
classification systems design, and mobile device based machine learning algorithm implementation.
• Proficient with mainstream modern deep learning packages (e.g. Tensorflow, Caffe and cuda-convnet), programming tools
(e.g. CUDA, OpenCV, OpenGL and OpenCL) and Unity simulation.
ACADEMIC BACKGROUND
• Ph.D. in Computer Science (GPA: 3.84)• Michigan State University, USA • 2012-present
§ Advisor: Juyang (John) Weng
§ Specialization: Cognitive Science
§ Research Interests: Artificial Intelligence, Machine Learning, and Neural Networks
§ Expected graduation date: April, 2017
• Bachelor of Science in Mathematics and Applied Mathematics • Fudan University, China • 2008-2012
§ Advisor: Jin Cheng
§ Dissertation: Convergence Properties of the Algebraic Reconstruction Technique and its Application
RESEARCH/SCHOLARLY ACTIVITIES
PUBLICATIONS
• Zheng, Z. and Weng, J. “Mobile Device Based Outdoor Navigation With On-line Learning Neural Network: a Comparison
with Convolutional Neural Network”. IEEE Conf on Computer Vision and Pattern Recognition Workshop (CVPRW). June 2016.
• Zheng, Z. and Weng, J. “Challenges in Visual Parking and How a Developmental Network Approaches the Problem”.
International Joint Conference on Neural Networks (IJCNN), July 2016.
• Zheng, Z., He, X, and Weng, J. “Approaching Camera-based Real-World Navigation Using Object Recognition”. INNS
BigData International Conference, Aug 2015.
• Zheng, Z., Li, Z., Nagar, A. and Park, K. “Compact Deep Neural Networks for Device Based Image Classification”. IEEE Int’l
Conf on Multimedia & Expo Workshop (ICMEW), July 2015.
• Zheng, Z., Li, Z., Nagar, A. “Compact Deep Neural Networks for Device Based Image Classification”. Mobile Cloud Visual
Media Computing: From Interaction to Service: 201-217, Springer, July 2015.
• Zheng, Z., and Weng, J. “Approaching Real-World Navigation Using Object Recognition Network”. International Joint
Conference on Neural Networks (IJCNN), June 2015.
• Zheng, Z., Weng, J., & Zhang, Z. WWN: Integration with coarse-to-fine, supervised and reinforcement learning. In Neural
Networks (IJCNN), 2014 International Joint Conference on (pp. 1517-1524). IEEE. July 2014.
• Zheng, Z., Qian, K., Weng, J., & Zhang, Z. Modeling the effects of neuromodulation on internal brain areas: Serotonin and
dopamine. In Neural Networks (IJCNN), The 2013 International Joint Conference on (pp. 1-8). IEEE. August 2013.
• Zheng, Z. & Weng, J. “Comparison between WWN and Some Prior Networks” International Conference on Brian Mind (ICBM)
June 2013.
• Zheng, Z., Kui, Q., Weng, J. & Zhang, Z. “Reinforcement in Internal Areas, Coarse-to-Fine Motors, and Integration with
Supervised Learning” Transaction on Autonomous Mental Development (Working paper).
Z E J I A Z H E N G
1870 E Shore Dr., Apt C2
East Lansing, MI 48823
Tel.: 517.703.4794
Email: zhengzej@msu.edu
Blog: zejiazheng.com
2. RESEARCH EXPERIENCE
• ZOOX (Stealth mode robotics startup) • Summer Intern • Supervisor: Paulius Micikevicius • 2016.5-2016.9
§ Worked with Zoox Computer Vision Group, lead by James Philbin.
§ Designed and programmed Zoox Deep Learning Inference Engine for autonomous driving systems.
§ Studied GPU memory allocation and deallocation mechanisms for pre-trained deep learning neural networks;
Inference Engine is currently being used as a standard for real-time neural network deployment on Zoox vehicles; it
saves more than 40% GPU memory required to run inference with modern deep neural network architectures (e.g.
VGG variants, Inception net, etc.) compared to Caffe’s approach.
• Samsung Research America • Summer Intern • Advisor: Dr. Zhu Li • 2014.5-2014.8
§ Worked with Compact Descriptor for Visual Search Group.
§ Constructed compact deep convolutional neural network for mobile device based image recognition.
§ Minimized memory footprint of device based image classifier by 32% compared to typical convolutional neural
network; developed efficient contribution evaluation method for classifier kernels; network recognizes single image
within 0.1 sec on mobile devices.
• Embodied Intelligence Laboratory • Michigan State University • Advisor: Dr. Juyang (John) Weng • 2012-present
§ Designed online learning mobile agent that learns to navigate in real-world environment with real-time attention
modeling.
§ Programmed Android application for outdoor navigation. The system navigates successfully around campus using
built-in camera and GPS, with the ability of direction correction and obstacles avoidance. System learns to recognize
objects and transfer learned concepts to unfamiliar environments.
§ Constructed reinforcement learning modules under the framework of Developmental Network; simulated the effect
of neurotransmitters in inner brain areas; network learns to navigate in simple maze without human supervision.
§ Utilized an instructional scaffolding learning scheme to train multi-concept learning agents from single concept
learning agent; reduced manual labeling in supervision by over 80%; Agent learns to refine concept of location
according to its interaction with external environment.
§ Research supported by Microsoft Research, Redmond
• Information Science Laboratory • Fudan University • Advisor: Dr. Peizhong Lu • 2011-2012
§ Developed novel clock synchronization algorithm for wireless sensor networks.
§ Applied parameter estimation techniques to large time stamp data collected from wireless sensor network.
§ Constructed simulation platform for pairwise synchronization and network wide synchronization.
• Research Center of Nonlinear Sciences • Fudan University • Advisor: Dr. Wei Lin • 2010-2011
§ Evaluated patient diagnosis report using statistical correlation and time series analysis models.
§ Analyzed Hemodialysis results with belief propagation network.
§ Collected data to improve evaluation results at Huashan Hospital, Shanghai.
PROFESSIONAL HONORS AND AWARDS
• Second Prize at China Undergraduate Mathematical Contest in Modeling • Paper: Pollution Analysis and Prediction Based
on Neural Networks • Sep 2012
• Outstanding Undergrad Scholarship of Fudan University • 2011-2012
• Second Prize at Huadong Cup, the 12th Mathematical Modeling Contest for Undergraduates (ranked 4th out of 640 teams) •
Paper: Mathematical Analysis and Modeling for Elevator Scheduling • Apr 2010
• Fudan Scholarship for Fundamental Sciences • 2009-2011
RELEVANT EXPERIENCE
CONTRACTED STATISTICIAN
Geosun Illumination Co. – Ningbo, China Jan 2012-May 2012
• Designed and led construction of long term employee performance database using SQL for 200 sales representatives.
• Developed a novel employee long term performance rating system through analysis of the ELO chess-player rating system.
TEACHING ASSISTANT
Computer Science and Engineering- Michigan State University SS2015, FS2014 and SS2014
• CSE 331 Data Structure and Algorithm. Designed programming projects for the course. Grading and office hour instructions
and support.
• CSE 260 Discrete Mathematics. Lecture and discussion for undergraduate students.
Z E J I A Z H E N G Page 2
3. TECHNICAL SKILLS
• Programming: • C/C++ • Matlab • Python • JAVA • Android • Unity
• Libraries/Tools: • LATEX • Qt4 • Matlab/Python/JAVA GUI Design • CUDA programming • OpenGL • OpenCV
REFERENCES
• Dr. Juyang (John) Weng, Computer Science and Engineering • Michigan State University, East Lansing, MI 48824 • phone:
517.353.4388 • email: weng@cse.msu.edu
• Dr. Fathi Salem, Electrical and Computer Engineering • Michigan State University, East Lansing, MI 48824 • phone:
517.355.7695 • email: salem@egr.msu.edu
• Dr. Eric Torng, Computer Science and Engineering • Michigan State University, East Lansing, MI 48824 • phone: 517.353.3543
• email: torng@cse.msu.edu
• Dr. Zhu Li, Compact Descriptor for Visual Search Group • Samsung Research America, Richardson, TX 75082 • email:
zhu1.li@samsung.com
• Dr. Wei Lin, School of Mathematical Science • Fudan University, Shanghai, China 200433 • phone: +86.021.55665141 • email:
wlin@fudan.edu.cn
Z E J I A Z H E N G Page 3