About

My name is Z. Sun, a 2nd year PhD student in Statistics at University College London. I am very very lucky to be supervised by Prof. F-X Briol and Prof. J-H Xue. Before that, I accomplished my M.Sc at University of Oxford.

Research Interest

In general, I am very interested in computational statistics and machine learning. Recently, I have worked on meta-learning, control variates and Stein’s method.

  • Meta-learning targets at the learning problems when only a small set of samples are available for training. It is all about how to make use of past knowledge to generalize on novel unseen tasks.

  • Control Variates methdods aim to get an estimator for integration problems which largely reduce the variance of Monte Carlo estimators especially when sample size is small.

Publication

[* means equal contribution.]

  • Sun, Z., Wu, J., Li, X., Yang, W., Xue, J-H.(2021). Amortized Bayesian Prototype Meta-learning: A new probabilistic meta-learning approach to few-shot image classification. In Proceedings of the 24th International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR.

  • Li, X.*, Wu, J.*, Sun, Z.*, Ma, Z., Cao, J., Xue, J-H.(2020). Bi-Similarity Network for Fine-grained Few-shot Image Classification. IEEE Transactions on Image Processing.

  • Li, X.*, Sun, Z.*, Xue, J-H., Ma, Z. (2020). A Concise Review of Recent Few-shot Meta-learning Methods. Neurocomputing.

Talks/Presentations

  • Poster at AISTATS 2021, April 2021, (San Diego, California, USA).
  • Talk at Data Centric Engineering @ The Alan Turing Institute, March 2021, (London, UK).