Facebook AI Research Overview

Category: Machine Intelligence

Read the original document

<!-- gdoc-inlined -->

1. Applications

1. Speech Recognition
   1. Learning Filterbanks from Raw Speech for Phone Recognition
      1. Neil Zeghidour, Nicolas Usunier, Iasonas Kokkinos, Thomas Schatz, Gabriel Synnaeve, Emmanuel Dupoux
2. Natural Language Understanding
   1. Word Embeddings
      1. Advances in Pre-Training Distributed Word Representations
         1. Tomas Mikolov, Edouard Grave, Piotr Bojanowski, Christian Puhrsch, Armand Joulin
   2. Language Modeling
      1. Unbounded Cache Model for Online Language Modeling with Open Vocabulary
         1. Edouard Grave, Moustapha Cisse, Armand Joulin
   3. StarSpace: Embed All the Things!
      1. Ledell Wu, Adam Fisch, Sumit Chopra, Keith Adams, Antoine Bordes, Jason Weston
3. Multi-Modal Learning
   1. Efficient Large-Scale Multi-Modal Classification
      1. Douwe Kiela, Edouard Grave, Armand Joulin, Tomas Mikolov
4. Multi-Agent
   1. VAIN: Attentional Multi-agent Predictive Modeling [Also, Attention]
      1. Yedid Hoshen
5. Attention
   1. Attentive Explanations: Justifying Decisions and Pointing to the Evidence [Also, Multi-Modal, Question Answering]
      1. Dong Huk Park, Lisa Anne Hendricks, Zeynep Akata, Anna Rohrbach, Bernt Schiele, Trevor Darrell, Marcus Rohrbach
6. Memory
   1. Gradient Episodic Memory for Continual Learning
      1. David Lopez-Paz, Marc’Aurelio Ranzato
7. Transfer Learning
   1. Best of Both Worlds: Transferring Knowledge from Discriminative Learning to a Generative Visual Dialog Model [Also, Generative]
      1. Jiasen Lu, Anitha Kannan, Jianwei Yang, Devi Parikh, Dhruv Batra
8. Representation Learning
   1. Poincare Embeddings for Learning Hierarchical Representations
      1. Maximilian Nickel, Douwe Kiela
   2. Fader Networks: Manipulating Images by Sliding Attributes
      1. Guillaume Lample, Neil Zeghidour, Nicolas Usunier, Antoine Bordes, Ludovic Denoyer, Marc’Aurelio Ranzato
         1. 9. Adversarial Examples
   3. Houdini: Fooling Deep Structured Visual and Speech Recognition Models with Adversarial Examples
      1. Moustappha Cisse, Yossi Adi, Natalia Neverova, Joseph Keshet
9. Datasets, Environments
10. ELF: An Extensive, Lightweight and Flexible Research Platform for Real-time Strategy Games
    1. Yuandong Tian, Qucheng Gong, Wenling Shang, Yuxin Wu, Larry Zitnick
11. Unsupervised Learning
12. On the Optimization Landscape of Tensor Decompositions
    1. Rong Ge, Tengyu Ma

99 People on their ‘people’ page.

Breakdown from Peter:

FAIR has 120 people, ~40 in mtv, paris and new york FAIR has ~30 software engineers, 16 of which work on Pytorch Leaves about 90 research scientists

Source: Original Google Doc