Argmax

2: data2vec

Vahe Hagopian, Taka Hasegawa, Farrukh Rahman Season 1 Episode 2

Todays paper: data2vec (https://arxiv.org/abs/2202.03555)

Summary of the paper
A multimodal SSL algorithm that predicts latent representation of different types of input.

Highlights of discussion

  • What are the motivations of SSL and multimodal
  • How does the student teacher learning work?
  • What are similarities and differences between ViT, BYOL, and Reinforcement Learning algorithms.