Ryan Gao
Education
2013.09 - Now
Jiangnan University - Mircoelectronics Bachelor
2016.12 - Now
Udacity - Self-Driving Car Nanodegree
2016.07 - 2016.11
Udacity - Machine Learning Nanodegree
Projects
2017.4-Now
Object Detection: Multi-View 3D object Detection - github.com/RyannnG/MV3D_TF(in progress)
● Using Lidar and Mono camera to detect objects’ 3D locations
● Re-implemented Faster-RCNN on Tensorflow
2016.12
Computer Vision: Advanced Lane Detection - github.com/RyannnG/CarND-Advanced-Lane-Lines
● Using color transform and gradients to create thretholded binary images
● Finding lane pixels and fit to find lane curvature
2016.11
Computer Vision: Google SVHN Recognition- github.com/RyannnG/Capstone-Google-SVHN-Digits-Recognition
● Built a small ConvNet on Tensorflow which capable of recognizing number sequnces
2016.10
Q-Learning: Teaching car how to drive- github.com/RyannnG/Udacity-Machine-Learning-Nanodegree
● Applied Q-learning for a self-driving agent in a game world to reach its destinations in the allotted time.
● Improved upon the Q-Learning algorithm to find the best configuration of learning and exploration factors to ensure the self-driving agent was reaching its destinations
Internships
2017.02 – Now
Uisee Technology (Beijing) Ltd – Computer Vision Intern
● 3D object detection using stereo image pairs
● 3D object detection using Lidar and mono images
2016.07 – 2016.10
Variable Supercomputer Tech – Algorithm Intern
● Collected leased tokens on the Internet, and abstracted them from the raw data
● Analyzed the leased tokens to get their statistical characteristics using Python
● Participated in testing the computing nodes.
Skills
Language:Python, C/C++
Tools: Numpy, Tensorflow, OpenCV
Others: Git/GitHub
教育经历
2013.09 - 2017.06
江南大学 - 微电子科学与工程, 本科
2016.07 - 2016.11
Udacity - 机器学习纳米学位
2016.12 - 至今
Udacity - 无人驾驶纳米学位
项目经历
2016.12
探测车道线 - github.com/RyannnG/CarND-LaneLines-P1
● 对路况图片实施Canny变换检测出边缘, 随后施加Hough变换画出车道线
2016.11
ConvNet识别SVHN数字- github.com/RyannnG/Capstone-Google-SVHN-Digits-Recognition
● 利用Tensorflow构建5层ConvNet模型(限于电脑资源),可用于识别5位数字串
● 数字串识别准确率74.5%, 单个数字识别准确率率92.3%
2016.10
智能游戏小车- github.com/RyannnG/Udacity-Machine-Learning-Nanodegree
● 依据游戏情景设置状态表,设置Q-Learning算法让自动行驶的小车在规定时间内到达目的地
● 调整Q-Learning 算法参数,找到合适的学习参数并保证小车持续接受正向奖励(遵守交通规则)
2016.08
建立学生监督系统- github.com/RyannnG/Udacity-Machine-Learning-Nanodegree
● 统计分析学生过往毕业信息,建立并比较不同监督学习模型(决策树/SVM/kNN)
● 预测给定学生是否可以顺利毕业,以确定是否需要学校提前采取措施
实习经历
2016.07 – 2016.10
江苏微锐超算科技有限公司 – 算法实习生
● 收集并提取网上泄露口令
● 用Python统计分析口令特征
● 参与计算节点测试
工具技能
编程语言:Python(熟练), C/C++(基础)
库: Scikit-Learn, Tensorflow, OpenCV
其他: Git/GitHub
联系方式
邮箱 :gy.shouwang@gmail.com
Github: github.com/RyannnG
领英: linkedin.com/in/ryangaoyuan
欢迎联络😎