田品卓 Working on Ph.D. degree Department of Computer Science and Technology Nanjing University E-mail: tianpinzhuo [@] gmail [DOT] com

About me

I'm interested in machine learning. Recently, I mainly focus on how to use meta-learning or transfer learning to design efficient algorithms. These algorithms are supposed to learn useful knowledge with a few training data or under some limitations. I'm supervised by Prof. Gao Yang.

News

• [Jun. 2021] One paper is accepted by FCS!

• [Jun. 2021] One paper is accepted by TNNLS!

Publications

Journal

1. Pinzhuo Tian, Wenbin Li, Yang Gao, "Consistent Meta Regularization for Better Meta-knowledge in Few-shot Learning ",  TNNLS 21,  (to appear).

2. Pinzhuo Tian, Yang Gao, "Improving Meta-Learning Model via Meta-contrastive Loss ",  Frontiers of Computer Science (FCS) 21,  (to appear).

Conference

1. Pinzhuo Tian, Lei Qi, Shaokang Dong, Yinghua Shi, Yang Gao, "Consistent MetaReg: Alleviating Intra-task Discrepancy for Better Meta-knowledge ",  IJCAI 20,  [PDF].

2. Pinzhuo Tian, Zhangkai Wu, Lei Qi, Lei Wang, Yinghuan Shi, Yang Gao, " Differentiable Meta-Learning Model for Few-Shot Semantic Segmentation ",  AAAI 20,  [PDF].

3. Pinzhuo Tian, Lei Qi, Yinghuan Shi, Luping Zhou, Yang Gao, Dinggang Shen, " A Novel Image-Specific Transfer Approach for Prostate Segmentation in MR Images ",  ICASSP 18  [PDF].

Awards

• Scholarship of Artificial Intelligence (2020)

• Outstanding Award of Ph.D. students (2020)

• Academic Exchange Award of Ph.D. students (2018)