Selected Skills

Python

tensorflow/sci-kit/opencv/flask

Statistics

numpy/pandas/matplotlib/seaborn

Prototyping

PS/AE/FCPX/XD/Sketch

Software Development

vue/electron/SSM/MySQL
Redis/ES/Kafka/Docker

Teamwork

Lean/Kanban/Time-Management

Languages

Native Chinese/Fluent English
Python/Java/JavaScript/Swift

Work Experience

 
 
 
 
 
2022-01 – Present
Suzhou, China

Software Engineer

Microsoft (China) Co., Ltd.

Deliver high-quality product.
 
 
 
 
 
2020-07 – 2022-01
Hangzhou, China

System R&D Engineer

CMB

  1. Focus on bank wholesale business analysis, CRM system backend design and development;
  2. System performance optimizing;
  3. And much more.
 
 
 
 
 
2019-07 – 2019-09
Hangzhou, China

NLP Algorithm Intern

DXY

  1. Combined with the business, to understand the algorithm requirements, and complete the design and implementation of relevant algorithms.
  2. Combined with natural language processing and machine learning methods, to conduct query understanding and user intention analysis, correlation score calculation and sorting optimization.
  3. Machine reading comprehension research in the domain of professional knowledge.

Experience

 
 
 
 
 
May 2019 – June 2019
Chengdu, China

The audio-semantic model for cognitive development of robots

First Author

Propose an audio-semantic model for cognitive development of robot, which makes the robot can dance according to the emotions expressed by the music and the memory related to the music. The paper has been published on the conference ICONIP2019
 
 
 
 
 
January 2019 – March 2019
Chengdu, China

Research on the QA problem in NLP modeling based on meta-learning

Second Author

We propose QA problem model based on meta-learning with small samples and simple structure. Participate in the research of a meta-learning-based efficient question-and-answer model that yields better or competitive results compared with the MemNN, EntNet and MemN2N model on the 1k bAbI dataset and the scaled down CBT dataset. The paper has been submitted to the conference ACL2019.
 
 
 
 
 
October 2018 – February 2019
Chengdu, China

Research on the associative memory model for cognitive development of robots

First Author

Lead the research of bidirectional associative memory (BAM) model based on spiking neural networks, and build the system for vision-action semantic associative learning of NAO robot.
 
 
 
 
 
June 2018 – September 2018
Chengdu, China

The second national collegiate contest for brain-inspired computing

Team Member

The language cognitive model based on Spiking Neural Network and Entity Network is proposed, and trying to solve the QA problem in NLP with SNN. Finally we win the second prize in the preliminary contest. (32⁄400)
 
 
 
 
 
April 2018 – June 2018
Chengdu, China

Research on gesture recognition method based on SNN

Second Author

We propose the gesture recognition method based on spiking neural networks, the paper “A Gesture Recognition Method Based on Spiking Neural Networks for Cognition Development” has been published on the conference ICONIP2018.
 
 
 
 
 
July 2017 – September 2017
Chengdu, China

The 3rd Sichuan ”Internet +” college students innovation competition

Actual Lead

Based on the laboratory innovative results, the brain-inspired GPS chip is proposed for robot cognitive mapping and navigation. I give the road show speech and we win the silver award.

Accomplishments

October 2019 – October 2019

Python and Statistics for Financial Analysis

The Hong Kong University of Science and Technology (Coursera)

Course link here.
“Python is now becoming the number 1 programming language for data science. Due to python’s simplicity and high readability, it is gaining its importance in the financial industry. The course combines both python coding and statistical concepts and applies into analyzing financial data, such as stock data. “
January 2019

CS224n: Natural Language Processing with Deep Learning (2017)

Stanford

Course link here.
“In this course, students will gain a thorough introduction to cutting-edge research in Deep Learning for NLP. Through lectures, assignments and a final project, students will learn the necessary skills to design, implement, and understand their own neural network models. “
October 2018 – November 2018

Machine Learning Crash Course

Google

Course link here.
Machine Learning Crash Course features a series of lessons with video lectures, real-world case studies, and hands-on practice exercises.
April 2018 – June 2018

Deep Learning Specialization

Deeplearning.ai

Course link here.
The specialization includes 5 courses:
1. Neural Networks and Deep Learning
2. Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization
3. Structuring Machine Learning Projects
4. Convolutional Neural Networks
5. Sequence Models

Selected Projects

*
All Models Apps Designs Gadgets Other

ReDeQ: Redis Delayed Queue

Yet Another Redis Delayed Queue.

WeChat-iOS15 Mock Icon

An ios-15 wechat icon I guess

scu_thesis_2020

A non-official latex template for SCU thesis of MASTER/DOCTOR.

DJ-beats

Beats time markers generator for FCPX and PRE.

K-pic

A simple image cloud storage GUI uploader.

SunIsland 404 Page

A chill 404 page with parallel effects.

Visual-semantic model

Reproduction for paper about visual-semantic attractor networks.

NCRC Official website(offline)

The official website of neuromorphic computing research center. Developed with vue.js, SSM framework and RESTful style.

Selected Publications

In ICONIP

This work inspired by the functional structure of Limbic system of human brain constructs a system for cognitive development of robots, which based on the proposed model of audio mapping to semantics, to realize that robots can evoke emotions and related memories from music. Moreover, behavioral decisions are made through the similarity comparison between music semantics and dance semantics in their bag-of-words vector representation, and then the robot will dance to the music as feedback.
Preprint PDF Project Poster Author page
In ICONIP

This paper proposes a gesture recognition method based on spiking neural network (SNN). The method can be used to develop the cognition behavior by associating the recognition results with semantic information from the observed target.
PDF Author page

Recent Publications

This work inspired by the functional structure of Limbic system of human brain constructs a system for cognitive development of robots, …
Preprint PDF Project Poster Author page

This paper proposes a gesture recognition method based on spiking neural network (SNN). The method can be used to develop the cognition …
PDF Author page

This is my personal notes for machine learning crash course from google.
PDF Project

Contact

  • College of Computer Science, Sichuan University. No. 24 South Section 1, Yihuan Road, Chengdu, Sichuan, China🇨🇳
  • Send me an E-mail