Fupin Yao
adjo96@r.postjobfree.com Guildford,UK,GU2 7XJ +44-747******* Personal website Github
Education
Master student in Computer Engineering Purdue University, West Lafayette, IN, USA Field of Research: Computer vision and deep learning 2011 Sep - 2016 June:
B.S. in Computer Sci. and Tech. Department Huazhong Univ. of Sci. and Tech., Wuhan, China Research Experience
Ph.D. Researcher, iFlyTek-Surrey Joint Research Centre on Artificial Intelligence,UK 2019 Oct - 2021 Jan
Proposed a new and realistic setting, cross-domain few-shot learning and designed a benchmark for this setting.
Proposed a contrastive learning based approach and a rotation based approach and our methods improve the accuracy from 60% (the baseline) to 70%.
Published two paper based on the above results.
Research assistant, Purdue University, USA 2018 Jan - 2019 Jan
Designed and implemented a new type of spherical convolutional layer, a new type of spherical pooling layer and convolutional neural networks for 3d models classification and retrieval.
Achieved superior accuracies over the state of the art and publish a paper in ICLR. Research assistant at NTU, Singapore 2016 July - 2016 Dec
Proposed a new research topic: grasping object in a way like humans
Designed and implemented a novel end-to-end convolutional neural network for object detection and grasping part detection and achieved high accuracy. The visual info is used for robot grasping.
Published a paper based on the above result.
Industrial Experience
Software engineer and CEO, Shulaishuwang Ltd., Wuhan, China. 2015 Apr - 2015 Oct
Worked as a android and back-end engineer and CEO of a 8-people team
Contributed a large portion of android client development and server development.
Released an android application for connecting people from the same hometown. Overview Download.
Software engineer and CTO, Baijinglianchuang Ltd., Wuhan, China. 2014 Oct - 2015 Apr
Worked as CTO in a start-up internet company and was in charge of a 10-engineer team.
Contributed a large portion of android client development, released an android application for college students, like Amazon. Overview Download.
Skills
Machine learning: Expert in few shot learning, meta learning, familiar with GNN, VAE, explainable AI, supervised learning, self-supervised learning, GNN, transformers. Math: Good at linear algebra, probability, calculus. Know stochastic process and optimization. Coding: Has several years experience as a software developer, released several android apps Proficient in PyTorch, Tensorflow, Keras, Java, python, PHP, Git, Docker, MySQL, Condor and Android. Familiar with C/C++, MatLab, OpenCV, QT, MFC, HTML, JavaScript, AWS, GCP and Colab Know hadoop, MapReduce
Language: Proficiency in English (IELTS 6.5 and TOFEL 92) and Mandarin (native) Publication
Min Liu, Fupin Yao, Chiho Choi, Sinha Ayan, Karthik Ramani. “Deep Learning 3D Shapes Using Alt- azimuth Anisotropic 2-Sphere Convolution” 7th International Conference on Learning Representations. (Top conference in deep learning)
Thalmann, Nadia Magnenat, Li Tian, and Fupin Yao. "Nadine: A Social Robot that Can Localize Objects and Grasp Them in a Human Way." Frontiers in Electronic Technologies. Springer, Singapore, 2017. 1-23. (Responsible for object-recognition part)
Fupin Yao “Cross-domain Few-shot Learning with Unlabelled Data” Arxiv, pre-print.
Fupin Yao “Machine learning with limited data” Arxiv, pre-print. Awards
Recipient of the Student Travel Grant of IEEE Automatic Face and Gesture Recognition(2017) Second Prize in Huawei Cup intelligent design contest(2014) Third Prize in China Internet business plan contest(2013) 1985 Alumni scholarship,HUST(2012)
Outstanding Student Scholarship, HUST (October, 2012)