Tim Z. Xiao

Hello! I am Tim, also known as Zhenzhong Xiao (肖镇中 in Chinese). Currently I am doing my PhD in machine learning at the University of Tübingen with Robert Bamler.

Previously, I did a MRes Computational Statistics and Machine Learning at UCL with David Barber and a MSc in Computer Science at the University of Oxford with Yarin Gal. Prior to that I studied computer science at the University of Manchester for my undergraduate. At the time, I also had an one year industrial placement with Morgan Stanley, during which I worked with a global team to create tools for managing and monitoring the cloud platform that has a huge number of servers located around the world.

Just like lots of computer scientists, I have faith in technology. Even though technology won’t solve all the problems for you, I believe it will certainly take you closer to the answers.

At the moment, I am very interested in Machine Learning, a field that has amazing impacts on human life.

news

Nov 3, 2023	I was invited to give a talk at Imperial College London - CSML reading group!
Nov 2, 2023	Check out our new workshop paper “The SVHN Dataset Is Deceptive for Probabilistic Generative Models Due to a Distribution Mismatch” to know more about distribution mismatch in the SVHN dataset!
Mar 13, 2023	Check out our new blog post “Large Language Models Are Zero-Shot Problem Solvers — Just Like Modern Computers” to find out the interesting connection between LLMs and modern computers!
Jan 20, 2023	Our paper “Trading Information between Latents in Hierarchical Variational Autoencoders” has been accepted at ICLR!
Jan 20, 2023	Our paper “Iterative Teaching by Data Hallucination” has been accepted at AISTATS!

selected publications

A Compact Representation for Bayesian Neural Networks By Removing Permutation Symmetry

Tim Z. Xiao, Weiyang Liu, and Robert Bamler

arXiv preprint arXiv:2401.00611, 2023

NeurIPS 2023 Workshop on Unifying Representations in Neural Models

arXiv Bib HTML PDF Poster

@article{xiao2023compact,
  title = {A Compact Representation for Bayesian Neural Networks By Removing Permutation Symmetry},
  author = {Xiao, Tim Z. and Liu, Weiyang and Bamler, Robert},
  journal = {arXiv preprint arXiv:2401.00611},
  year = {2023},
}

The SVHN Dataset Is Deceptive for Probabilistic Generative Models Due to a Distribution Mismatch

Tim Z. Xiao*, Johannes Zenn*, and Robert Bamler

arXiv preprint arXiv:2312.02168, 2023

NeurIPS 2023 Workshop on Distribution Shifts

arXiv Bib HTML PDF Poster

@article{xiao2023the,
  title = {The SVHN Dataset Is Deceptive for Probabilistic Generative Models Due to a Distribution Mismatch},
  author = {Xiao, Tim Z. and Zenn, Johannes and Bamler, Robert},
  journal = {arXiv preprint arXiv:2312.02168},
  year = {2023},
}

Trading Information between Latents in Hierarchical Variational Autoencoders

Tim Z. Xiao, and Robert Bamler

In International Conference on Learning Representations (ICLR), 2023

arXiv Bib HTML PDF Code Poster Slides

@inproceedings{xiao2023trading,
  title = {Trading Information between Latents in Hierarchical Variational Autoencoders},
  author = {Xiao, Tim Z. and Bamler, Robert},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2023},
}

Wat zei je? Detecting Out-of-Distribution Translations with Variational Transformers

Tim Z. Xiao, Aidan N. Gomez, and Yarin Gal

arXiv preprint arXiv:2006.08344, 2020

Spotlight talk, NeurIPS 2019 Workshop on Bayesian Deep Learning

Abs arXiv Bib HTML PDF Poster Slides

My master project at Oxford. In the project, we detect out-of-training-distribution sentences in Neural Machine Translation using the Bayesian Deep Learning equivalent of Transformer models. For this we develop a new measure of uncertainty designed specifically for long sequences of discrete random variables – i.e. words in the output sentence. Our new measure of uncertainty solves a major intractability in the naive application of existing approaches on long sentences. We use our new measure on a Transformer model trained with dropout approximate inference. On the task of German-English translation using WMT13 and Europarl, we show that with dropout uncertainty our measure is able to identify when Dutch source sentences, sentences which use the same word types as German, are given to the model instead of German.
@article{xiao2020wat, title = {Wat zei je? Detecting Out-of-Distribution Translations with Variational Transformers}, author = {Xiao, Tim Z. and Gomez, Aidan N. and Gal, Yarin}, journal = {arXiv preprint arXiv:2006.08344}, year = {2020}, }