Jure Žbontar | Barlow Twins: Self-Supervised Learning via Redundancy Reduction

Поділитися
Вставка
  • Опубліковано 17 лис 2024

КОМЕНТАРІ • 9

  • @coemgeincraobhach236
    @coemgeincraobhach236 2 роки тому

    Really appreciated this video, really interesting work. The frequent referral to other methods and how they differ was very clear and useful. Thanks!

  • @johntanchongmin
    @johntanchongmin 2 роки тому

    very nice explanation. I learnt a lot, thanks.

  • @joaobarreira5221
    @joaobarreira5221 3 роки тому

    hi,
    very nice job and very well explained.
    The subject of self-supervised learning always inspires me a feeling that I'm watching a sci-fi movie.
    It's so cool.

  • @buoyrina9669
    @buoyrina9669 2 роки тому

    Excellent talk

  • @paulsingerman2361
    @paulsingerman2361 Рік тому

    I do not understand. If this loss is calculated per image and simply averaged/summed over the batch how does this not just have the encoder collapse every single image to the same set of orthogonal vectors which would minimize the loss?

  • @anishbhanushali
    @anishbhanushali 3 роки тому

    very well explained

  • @clementatzberger38
    @clementatzberger38 3 роки тому +1

    Great work and nice illustration. However, I recommend that you change the naming of what you call "cross-correlation matrix" (in particular wrt the discussion starting at 20:00). What you optimize is not a correlation matrix in the proper sense (but just a matrix multiplication between ZA and ZB). The values in a real correlation matrix (between two matrices ZA and ZB being both hypothetically composed of 1's) would not be 1's but NaN because the covariance matrix is zero everywhere

    • @mushtaqml655
      @mushtaqml655 3 роки тому +1

      Can you elaborate that how they are not 1's? because according to the code he showed the cross-correlation matrix values will be divided by "N" which is the batch size, means normalization. Please correct me if I understood something wrong here. Thanks

    • @kazz811
      @kazz811 2 роки тому

      It is a matrix of correlations between specific features of the latent, averaged over the batch.