Gabriel Huang (he/him)

PhD, Mila & University of Montreal
Research Scientist, ServiceNow

Update March 2022: I am looking for a full-time position to pursue research on Responsible AI.
Topics of interest: impact of AI on human behavior and social dynamics, ethical and emotional AI, aligning AI with humanist values, self-supervised and low-data learning, multimodal transformers, and large language models.

Aligning AI with technical goals was a central theme of my PhD thesis, during which I have developed technical skills on multimodal language models, self-supervised and few-shot learning, object detection, optimal transport and generative learning. Now, I want to apply these technical skills to develop responsible and ethical AI aligned with humanist values, not only to ensure that individuals are treated fairly, but also to strengthen the core values of our societies (e.g. human life, truth, democracy, sustainability). I have recently started Naive Psychology, a blog to share my ideas on the interplay between AI, human behavior and society.

About me: I am a PhD student at Mila Quebec AI Institute advised by Simon Lacoste-Julien. Currently, I am a visiting researcher at ServiceNow (ElementAI) working on few-shot and self-supervised object detection, low-data language models, and self-supervised representations for climate change monitoring. Previously, I was an intern at Google Research working on multimodal pretraining for video captioning. I also hold two MSc. from Ecole Normale Supérieure and CentraleSupélec, where I specialized in machine learning, computer vision and statistics.

Publications

arxiv A Survey of Self-Supervised and Few-Shot Object Detection
Gabriel Huang, Issam Laradji, David Vazquez, Simon Lacoste-Julien, Pau Rodriguez
Submitted to IEEE TPAMI.

arXiv project page tweetorial Awesome-FSOD

paper Repurposing Pretrained Models for Robust Out-of-domain Few-Shot Learning
Namyeong Kwon, Hwidong Na, Gabriel Huang, Simon Lacoste-Julien
ICLR'21 paper.

arXiv pdf openreview

paper Multimodal Pretraining for Dense Video Captioning
Gabriel Huang, Bo Pang, Zhenhai Zhu, Clara Rivera, Radu Soricut
Introduces the Video Timeline Tags dataset (ViTT).
AACL-IJCNLP 2020.

arXiv dataset video slides

arXiv Are Few-Shot Learning Benchmarks too Simple ? Solving them without Task Supervision at Test-Time
This paper introduces Centroid Networks for Few-shot Clustering and Unsupervised Few-shot Classification
Gabriel Huang, Hugo Larochelle, Simon Lacoste-Julien
ICLR'19 workshop.

arXiv pdf code

paper Negative Momentum for Improved Game Dynamics
Gauthier Gidel, Reyhane Askari Hemmat, Mohammad Pezeshki, Gabriel Huang, Rémi Lepriol, Simon Lacoste-Julien, Ioannis Mitliagkas.
AISTATS 2019

arXiv slides visualization

arXiv Parametric Adversarial Divergences are Good Task Losses for Generative Modeling
Gabriel Huang, Hugo Berard, Ahmed Touati, Gauthier Gidel, Pascal Vincent, Simon Lacoste-Julien.
ICML'17 Workshop, ICLR'18 Workshop, Montreal AI Symposium 2018, Submitted to JMLR

arxiv pdf openreview

paper Scattering Networks for Hybrid Representation Learning
Edouard Oyallon, Sergey Zagoruyko, Gabriel Huang, Nikos Komodakis, Simon Lacoste-Julien, Matthew Blaschko, Eugene Belilovsky.
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) 2018

paper

Negative Momentum

Below is an interactive visualization of our paper Negative Momentum for Improved Game Dynamics:
(a) Learning rate (lr) and momentum (beta) hyperparameters.
(b) Resulting eigenvalues in the complex plane for SGD and SGD+momentum.

There is convergence if and only if all eigenvalues are inside the convergence ball (green).
Try to find the hyperparameters for convergence.

SGD without momentum: using , eigenvalues are the convergence ball →
SGD with momentum: using and momentum , eigenvalues are the convergence ball →

Thin-8 dataset

The Thin-8 dataset consists of 1585 grayscale handwritten images of the digit 8, with resolution 512x512.
16 people were asked to draw the digit 8 about 100 times using a pen on a tablet PC running Microsoft Windows.
It was collected in October 2017 at the University of Montreal.
Download Thin-8 dataset here

If you use the Thin-8 dataset, please cite our paper :

@article{huang2018parametric,
                    title={Parametric Adversarial Divergences are Good Task Losses for Generative Modeling},
                    author={Huang, Gabriel and Berard, Hugo and Touati, Ahmed and Gidel, Gauthier and Vincent, Pascal and Lacoste-Julien, Simon},
                    journal={arXiv preprint arXiv:1708.02511},
                    year={2017}
                  }

Thanks to Alex, Akram, Aristide, David, Dendi, Eugene, Jae, Joao, Liam, Rémi, Rosemary, Shawn, Sina, and Xing for scribbling all those samples!

Twitter Feed

Contact

Email: gbxhuang@gmail.com
In person: Mila, 6666 St-Urbain, #200, Montreal, QC, H2S 3H1, Canada