Who am I?

Welcome! Here is my Resume(PDF)

Academic

I’m an electrical and electronic engineer turned artificial intelligence engineer who loves investigate pattern recognition and multi-agent reinforcement learning. My latest project builds a passenger flow detection system at the metor hub. Right now, I’m working in a research lab to develop a dynamic control system by reinforcement learning.

Currently, I am a master student at Illinois Institute Technology (IIT) in the Computer Science department and applying for the Ph.D. program. I am interested in deep reinforcement learning in games, medical image recognition and machine learning with transparency.

Work Experience

Algorithm developer intern in Nise AI company (Dec 2017 to Mar 2019). I have done:
1. Developed monitoring algorithm for pedestrian flow depending on Mask-RCNN method by using Keras to help police monitor subway transportation.
2. Developed work attendance system by pattern recognition to improve employee sign in efficiency.
3. Developed python tool which could filter and mark image samples automatically by ResNet-101 in Keras framework. This tool helped our company reduced sample label makers form twelve to three.

Project Experience

Research assistant in Wave Lab(Expected to finish at May 2019). Developing a core understanding of critical capabilities and limitations of RL for inverse problems in science and engineering. Implement RL in a 2D system control problem, which leads to the data-driven embodiment of an optimal control scheme.
Research assistant in Pattern Recognition Lab (Jan 2016 to Nov 2017). I have done:
1. Modified the matching function from global training to additional dynamic training of Siamese INstance search Tracker(SINT) in visual tracking. Increased car tracking accuracy from 66.68% to 69.47%
2. Modified the Very Deep Convolutional Neural Network(VGG16) architecture by adding two performance in fer2013 (Public human facial expression database). Defeated Kaggle champions in 2013.
3. Identified the vehicle license plate within various noises (Such as Gaussian noise, image deletion, and image dithering.) condition by using MATLAB.
Facial Expression Recognition (Personal Project) (Feb 2015 to Jun 2015). Applied pattern recognition to implement human facial expression recognition. Used standard Japanese women face expression database (Jaffe). Combined Gabor filter and LBP as feature extraction, then using PCA to reduce features. Finally, classified features by LIBSVM.

Publication

H. Jiang, K. Huang, T. Wu, R. Zhang, T.O. Ting, And C. Wang: Robust One-Shot Facial Expression Recognition with Sunglasses, International Journal of Machine Learning and Computing, vol.6, No.2, pp. 80-86, April2016.

Competition

[C++] Robot Development Competition (Oct 2014 to Dec 2014). Developed and assembled the robot kits (Micro controller programming). Implemented various functions: The ambient light detection, Buzzer singing, Automatic tracing, Automatic avoid obstacles and Bluetooth control. Won the first prize of the Xi’an Jiaotong-liverpool University robot development competition.
[Android SDK] Software Design Competition (Sep 2014 to Nov 2014). Developed a smartphone application which enables multiple clients to make a group clock. Implemented overall Android application logic and wrote API server for community service. Won the third prize of the Xi’an Jiaotong-liverpool University software design competition.

Oneline Course

Deep reinforcement learning Nanodegree (Udacity)
Convolutional Neural Networks (Coursera)
Structuring Machine Learning Projects (Coursera)
Neural Networks and Deeplearning (Coursera)
Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization (Coursera)
Machine Learning (Coursera)
Python Data Structures (Coursera)
Using Python to Access Web Data (Coursera)

Programming Skill

Languages: Python, Jave, C++, Latex, Matlab, R
Tools: Tensorflow, Keras, Sklearn, Pygames, Version Control, Weka, Hugin Lite, Jekyll