Tatjana Chavdarova
Postdoctoral Researcher in ML
@ UC Berkeley, California

I am a Postdoctoral Researcher at the Department of Electrical Engineering and Computer Science (EECS) at the University of California at Berkeley, working with Michael Jordan. My research is supported by the Swiss National Science Foundation.

Prior to my current position, I was Postdoctoral Research Scientist at the Machine Learning and Optimization (MLO) lab at EPFL, working with Martin Jaggi. While at EPFL, I organized the Smooth Games reading group . In part of my time, I participated in the intelligent Global Health (iGH) sub-group of MLO led by Mary-Anne Hartley, by advising on the machine learning aspect of the ongoing projects.

I obtained my Ph.D. from EPFL, and Idiap, supervised by François Fleuret. During my Ph.D. studies I did two internships at: (i) Mila where I was supervised by Yoshua Bengio and Simon Lacoste-Julien, as well as at (ii) DeepMind supervised by Irina Higgins.

Research interests

My main interests lie at the intersection of game theory and machine learning. I would like to understand the training dynamics of multi-player games and saddle point optimization. In addition, my interests include unsupervised learning, generative modeling, generalization, and robustness.

Selected Publications and Preprints

• (workshop) Last-Iterate Convergence of Saddle Point Optimizers via High-Resolution Differential Equations
  Tatjana Chavdarova*, Michael I. Jordan*, Manolis Zampetakis*
  OPT NeurIPS workshop, 2021 • Paper • ArXiv • Poster • BibTex @article{chavdarova2021hrdes, title = {Last-Iterate Convergence of Saddle Point Optimizers via High-Resolution Differential Equations}, author = {Tatjana Chavdarova and Michael I. Jordan and Manolis Zampetakis}, booktitle= {NeurIPS Workshop on Optimization for Machine Learning}, year = {2021}, }
• (workshop) The Peril of Popular Deep Learning Uncertainty Estimation Methods
  Yehao Liu*, Matteo Pagliardini*, Tatjana Chavdarova, Sebastian U. Stich
  BDL NeurIPS workshop, 2021 • Paper • Code • Poster • BibTex @article{liu2021peril, title = {The Peril of Popular Deep Learning Uncertainty Estimation Methods}, author = {Yehao Liu and Matteo Pagliardini and Tatjana Chavdarova and Sebastian U. Stich}, booktitle= {NeurIPS Workshop on Bayesian Deep Learning}, year = {2021}, }
• (workshop) Improved Adversarial Robustness via Uncertainty Targeted Attacks
  Gilberto Manunza*, Matteo Pagliardini*, Martin Jaggi, Tatjana Chavdarova
  UDL ICML workshop, 2021 • Paper • Poster • BibTex @inproceedings{ManunzaPagliardini2021, title = {Improved Adversarial Robustness via Uncertainty Targeted Attacks}, author = {Gilberto Manunza and Matteo Pagliardini and Martin Jaggi and Tatjana Chavdarova}, booktitle= {ICML Workshop on Uncertainty and Robustness in Deep Learning}, year = {2021}, url = {http://www.gatsby.ucl.ac.uk/~balaji/udl2021/accepted-papers/UDL2021-paper-096.pdf} }
• Semantic Perturbations with Normalizing Flows for Improved Generalization
  Oguz K Yüksel, Sebastian U. Stich, Martin Jaggi, Tatjana Chavdarova
  ICCV, 2021 • Paper • Code • Video • BibTex @InProceedings{Yuksel_2021_ICCV, author = {Y\"uksel, Oguz Kaan and Stich, Sebastian U. and Jaggi, Martin and Chavdarova, Tatjana}, title = {Semantic Perturbations With Normalizing Flows for Improved Generalization}, booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)}, month = {October}, year = {2021}, pages = {6619-6629} }
• Taming GANs with Lookahead-Minmax
  Tatjana Chavdarova*, Matteo Pagliardini*, Sebastian U. Stich, François Fleuret, Martin Jaggi
  ICLR 2021 •Paper •Code •Video •Poster •BibTex @inproceedings{chavdarova2021lagan, author = {Tatjana Chavdarova and Matteo Pagliardini and Sebastian U. Stich and Fran{\c{c}}ois Fleuret and Martin Jaggi}, title = {Taming GANs with Lookahead-Minmax}, booktitle = {{ICLR}}, publisher = {OpenReview.net}, year = {2021} }
• (PhD thesis) Deep Generative Models and Applications
  Tatjana Chavdarova, Jury: F. Fleuret, P. Frossard, L. Denoyer, S. Lacoste-Julien, M. Jaggi.
  EPFL, July 2020 •Thesis •BibTex @article{Chavdarova:278463, title = {Deep Generative Models and Applications}, author = {Chavdarova, Tatjana}, institution = {IEL}, publisher = {EPFL}, address = {Lausanne}, pages = {169}, year = {2020}, url = {http://infoscience.epfl.ch/record/278463}, doi = {10.5075/epfl-thesis-10257}, }
• Reducing Noise in GAN Training with Variance Reduced Extragradient
  Tatjana Chavdarova*, Gauthier Gidel*, François Fleuret, Simon Lacoste-Julien
  NeurIPS 2019 •Paper •Code •Slides •Poster •BibTex @inproceedings{chavdarova2019, Author = {Tatjana Chavdarova and Gauthier Gidel and François Fleuret and Simon Lacoste-Julien}, Title = {Reducing Noise in {GAN} Training with Variance Reduced Extragradient}, Booktitle = {{Advances in Neural Information Processing Systems (NeurIPS)}}, Year = {2019}, volume = {32}, publisher = {Curran Associates, Inc.}, }
• SGAN: An Alternative Training of Generative Adversarial Networks
  Tatjana Chavdarova, François Fleuret
  CVPR 2018 •Paper •Slides •Poster •BibTex @inproceedings{chavdarova-fleuret-2018, author = {Chavdarova, T. and Fleuret, F.}, title = {{SGAN}: An Alternative Training of Generative Adversarial Networks}, booktitle = {CVPR}, year = {2018}, }
• WILDTRACK: A Multi-camera HD Dataset for Dense Unscripted Pedestrian Detection
  T. Chavdarova, P. Baqué, S. Bouquet, A. Maksai, T. Bagautdinov, L. Lettry, P. Fua, L. Van Gool, François Fleuret
  CVPR 2018 •Paper •Dataset •Code •Poster •BibTex @inproceedings{chavdarova-et-al-2018, author = {Chavdarova, T. and Baqué, P. and Bouquet, S. and Maksai, A. and Jose, C. and Bagautdinov, T. and Lettry, L. and Fua, P. and Van Gool, L. and Fleuret, F.}, title = {{WILDTRACK}: A Multi-camera {HD} Dataset for Dense Unscripted Pedestrian Detection}, booktitle = {CVPR}, year = {2018}, pages = {5030-5039}, }

* Equal contributions.

Talks and Presentations

• Generative Adversarial Networks
  Mediterranean Machine Learning (M2L) school, Jan. 2021 • video • slides • colab
• An Introduction to JAX
  Feb. 2020 • slides • colab
• Reducing Noise in GAN Training with Variance Reduced Extragradient
  Selected as oral presentation at Montreal AI Symposium and at ICML workshop on Generative Modeling and Model-Based Reasoning for Robotics and AI, 2019 • slides
• Generative Adversarial Networks: Algorithm, Challenges & Research Topics
  Paris GANs & co meetup, Oct. 2018 • slides
• SGAN: An Alternative Training of Generative Adversarial Networks & Research Topics
  Idiap/HES-SO AI workshop, April 2018 • slides

Supervising & Teaching

Other Services

Reviewing

• Conferences: • ICLR • ICML • AISTATS • NeurIPS
• Journals: • JMLR • SIAM Journal on Mathematics of Data Science

Workshops

• Program Chair. WiML Virtual Un-Workshop. Held at the International Conference on Machine Learning (ICML). July, 2020.
• Part of the Advisory Committee. ICCV Workshop on Multi-camera Multiple People Tracking, Oct., 2021.

Reading Group

• Organizer of the Smooth Games Reading Group at EPFL. Oct. 2020 - March 2021.

Misc

• Board of Directors for Women in Machine Learning. Since April 2021.

Contact

Feel free to reach out at: tatjana.chavdarova[at]berkeley[dot]edu