I joined
the University
of Geneva as Full Professor of CS.
I have five open phd positions at
the University of
Geneva . ↴
These are positions in the field
of deep learning for reinforcement learning, interpretability, density
models in large dimension, anomaly detection, and computational
reduction. Please apply through this online
form.
Our project “Robust Deep Density Models for High-Energy
Particle Physics and Solar Flare Analysis” is funded. ↴
This is a four year joint project
with Slava
Voloshynovskiy
and Tobias
Golling from
the University of
Geneva ,
and André
Csillaghy from
the University of Applied
Sciences and Arts Northwestern Switzerland. Funding comes
from
the Sinergia
program of the Swiss National Science Foundation.
Two papers accepted at ICML 2020. ↴
With Teja Prabhu, Florian Mai, Thijs Vogels, and Martin
Jaggi on the cost of hyper-parameter tuning
(Prabhu et
al, 2020 ).
With Angelos Katharopoulos, Apoorv Vyas, and Nikolaos Pappas
on speeding up transformers using a form of linear
attention
(Katharopoulos
et al, 2020 ).
My materials for EE-559 – “Deep Learning” are
available.
We organized
the 8th Swiss
Machine Learning
Day.
Two papers accepted at NeurIPS 2019. ↴
With Tatjana
Chavdarova , Gauthier
Gidel ,
and Simon
Lacoste-Julien on combining extra gradient and SVRG for
training GANs
(Chavdarova et al, 2019 ).
With Suraj
Srinivas on a gradient-based representation for deep
network visualization
(Srinivas
and Fleuret, 2019 ).
One paper accepted at ICCV 2019. ↴
With Stepan
Tulyakov , Martin Kiefel, Peter Gehler, and Michael
Hirsch on a new type of embedding to use deep stereo models
with event-based cameras
(Tulyakov et
al, 2019 ).
One paper accepted at ICML 2019. ↴
With
Angelos
Katharopoulos on using importance sampling to process
mega-pixel images with convolutional networks
(Katharopoulos
and Fleuret, 2019 ).
I taught deep learning at the AMMI in Kigali.
One paper accepted at NeurIPS 2018. ↴
With
Stepan
Tulyakov
and Anton
Ivanov on an efficient deep network and a new loss for
stereo reconstruction
(Tulyakov,
Ivanov, and Fleuret, 2018 ).
Four papers accepted at ICML 2018. ↴
With
Cijo Jose
and Moustafa
Cissé on Kronecker recurrent
units (Jose,
Cissé, and Fleuret, 2018 ).
With Suraj
Srinivas on deep
knowledge transfer with Jacobian-based
regularization (Srinivas
and Fleuret, 2018 ).
With Angelos
Katharopoulos on importance sampling for deep learning
(Katharopoulos
and Fleuret, 2018 ).
With Pierre
Baqué, Edoardo
Remelli,
and Pascal Fua on
shape optimization with deep
models (Baqué
et al, 2018 ).
Two papers accepted at CVPR 2018. ↴
With Tatjana
Chavdarova on a new strategy to train GANs using an
ensemble of pairs whose statistical independence is
maintained during
training (Chavdarova
and Fleuret, 2018 ).
With all the members of
the Wildtrack
project on a new multi-camera person tracking data-set
(Chavdarova
et al, 2018 ).
Best paper award at NIPS LLD workshop 2017. ↴
With Suraj
Srinivas , we got
the best
paper award at NIPS LLD workshop 2017 for a work
on deep
model transfer with linear approximations.
Best paper award at AISTATS 2017. ↴
With James Newling ,
we got
the Best
paper award at AISTATS 2017 for a work on a
sub-quadratic medoid algorithm
(Newling
and Fleuret, 2017 ). James also got
the Idiap
PhD student award.
One paper accepted at NIPS 2017. ↴
With James Newling
on k-means initialization
(Newling and
Fleuret, 2017b ).
Three papers accepted at ICCV 2017. ↴
On weakly supervised deep-learning for planet surface
reconstruction
(Tulyakov,
Ivanov and Fleuret, 2017 ).
On deep occlusion reasoning for people detection
(Baqué,
Fleuret and Fua, 2017b ).
On non-Markovian trajectory
modeling (Maksai et
al., 2017 )