Hi! I'm Thomas Robert

I am currently the R&D Team Lead & Deep Learning Research Engineer at Heuritech R&D Lab.

I defended my PhD on Deep Learning for Computer Vision from Sorbonne University (MLIA / LIP6) in October 2019.

A few things about me...

I love Computer Science

I started coding 11 years ago through web development and keep working on it has a hobby

I love Machine Learning

It's an incredible tool to solve complex problems

I love challenges

I am looking for innovative research & development projects using deep learning

I love learning and sharing

That's why I did a PhD and taught at my university

My main skills

Machine Learning & Data Science

  • Deep Learning
  • Data Science
  • PyTorch
  • TensorFlow
  • Keras
  • Matlab
  • Scikit-learn
  • Hadoop / Spark

Programming

  • Theoretical knowledge
  • Python
  • Java
  • Scala
  • C / C++
  • Bash
  • Git / SVN

Web

  • Theoretical knowledge
  • PHP
  • MySQL
  • JavaScript
  • HTML5
  • CSS3
Beginner / Proficient / Advanced or expert

See all my skills

Recent experiences

R&D Team Lead & Research engineer in Deep Learning

Heuritech – Paris, France

Nov. 2019 Current

Working in the R&D lab of Heuritech, a fashion tech company providing trends data to the fashion industry.

R&D Team Lead since June 2020.

PhD in Deep Learning

LIP6 lab, Sorbonne University (prev. Pierre and Marie Curie University) – Paris, France

2019 3 years

Improving Latent Representations of ConvNets for Visual Understanding - Slides

Jury:

Thesis Slides

Research engineer then PhD student in Deep Learning

LIP6 lab, Sorbonne University (prev. Pierre and Marie Curie University) – Paris, France

2019 4 years

Supervised by Matthieu Cord and Nicolas Thome

Work on the ANR projects VISIIR; et DeepVision in partnership with the LIRIS (INSA Lyon, France), the Simon Fraser University (Canada) and the University of Guelph (Canada).

  • Work on recipe image classification. Creation of a demonstration website of the classifier
  • Teaching work (64h / year): in charge of multiple courses, writing of convolutional networks practical sessions subjects
  • Submission to the workflow challenge of the M2CAI (MICCAI 2016) workshop with Remi Cadène : classification of surgical operation step based on endosopic videos. 2nd best model submitted.
  • Regularization of deep neural networks with Michael Blot. Published at ICIP 2018 (best paper award).
    SHADE: Information-Based Regularization for Deep Learning
  • Improvement of semi-supervised learning models based on auto-encoders. Published at ECCV 2018.
    HybridNet: Classification and Reconstruction Cooperation for Semi-Supervised Learning
  • Work on disentangling model for generation.
    DualDis: Dual-Branch Disentangling with Adversarial Learning

Publications and courses

Read more about my experiences

Research

Improving Latent Representations of ConvNets for Visual Understanding

T. Robert

PhD thesis, Sorbonne Université, Paris (2019) Jury: Stéphane Canu [Reviewer], Greg Mori [Reviewer], Catherine Achard [Examinator], Karteek Alahari [Examinator], David Picard [Examinator], Matthieu Cord [Supervisor], Nicolas Thome [Supervisor]

Thesis Slides

DualDis: Dual-Branch Disentangling with Adversarial Learning

T. Robert, N. Thome, M. Cord

arXiv (2019)

Paper

HybridNet: Classification and Reconstruction Cooperation for Semi-Supervised Learning

T. Robert, N. Thome, M. Cord

European Conference on Computer Vision (ECCV) (2018) Munich, Germany

Paper Poster

See all my publications