Thijs Vogels Pronunciation: /ˈtɛi̯s/

PhD student in Machine Learning & Optimization at EPFL
thijs.vogels@[epfl].ch

About

I am a PhD student in the Machine Learning and Optimization laboratory at EPFL under the supervision of Martin Jaggi. Before joining EPFL, I worked as a lab associate at Disney Research. I received a B.Sc. from University College Roosevelt, an honours college from Utrecht University, and a M.Sc. from ETH Zürich in Computational Science & Engineering.

Publications

Teaser for “Practical Low-Rank Communication Compression in Decentralized Deep Learning”

Practical Low-Rank Communication Compression in Decentralized Deep Learning

Published
Advances in Neural Information Processing Systems 33 (NeurIPS 2020)
Vancouver, Canada, December 7 – December 12, 2020.
Teaser for “Optimizer Benchmarking Needs to Account for Hyperparameter Tuning”

Optimizer Benchmarking Needs to Account for Hyperparameter Tuning

Published
Proceedings of the International Conference on Machine Learning (ICML 2020)
Vienna, Austria, July 13 – July 18, 2020.
Teaser for “PowerSGD: Practical Low-Rank Gradient Compression for Distributed Optimization”

PowerSGD: Practical Low-Rank Gradient Compression for Distributed Optimization

Published
Advances in Neural Information Processing Systems 32 (NeurIPS 2019)
Vancouver, Canada, December 8 – December 14, 2019.

Denoising with Kernel Prediction and Asymmetric Loss Functions

Published
ACM Transactions on Graphics (Proceedings of SIGGRAPH 2018), vol. 37, no. 4.
Vancouver, Canada, August 12 – August 16, 2018.
Teaser for “Kernel-predicting Convolutional Networks for Denoising Monte Carlo Renderings”

Kernel-predicting Convolutional Networks for Denoising Monte Carlo Renderings

Published
ACM Transactions on Graphics (Proceedings of SIGGRAPH 2017), vol. 36, no. 4.
Los Angeles, USA, July 05 – August 03, 2017.
Teaser for “Kernel-predicting Convolutional Neural Networks for Denoising Monte Carlo Renderings”

Kernel-predicting Convolutional Neural Networks for Denoising Monte Carlo Renderings

Published
Master’s thesis, ETH Zürich, 2016.
Teaser for “Web2Text: Deep Structured Boilerplate Removal”

Web2Text: Deep Structured Boilerplate Removal

Published
European Conference on Information Retrieval, 167–179.
Grenoble, France, March 26 – March 29, 2018.
Teaser for “Towards a Burglary Risk Profiler Using Demographic and Spatial Factors”

Towards a Burglary Risk Profiler Using Demographic and Spatial Factors

Published
Web Information Systems Engineering (WISE) 16, 586–600.
Miami, USA, November 1 – November 3, 2015.