Self Supervised Learning is Taking Off

 The cutting edge of deep learning research in self-supervised learning algorithms is taking off.  Self supervised being a neural net that learns from the data in the model, not from supervised labels or other pattern match-ups tagged by humans.  

And in some sense it's really all about data augmentation.  And we move from thinking about data augmentation as just a way to endlessly expand our model's data, to thinking about it as a way to introduce perceptual clustering of the data in our model so that the training process can manipulate it's energy surface to correspond to the natural perceptual classes in the data (doing this with human intervention).  Actually the human intervention is the human designing the correct perceptual augmentation.

SwAV is the latest paper from FAIR to generate state of the art results in self-supervised learning.  With exciting potential for an alternative approach to transfer learning (alternative to a ResNet model (or equivalent)).



Here's a more relaxed conversation with Mathilde Caron (paper author in the first presentation) on Machine Learning Street Talk.

Observations
1.  Yannic mentions how disappointing it is to not be able to load the full data set into memory to train the clustering.  But is that really true?  Stochastic gradient descent works in some sense because of the batch processing introducing randomization into the gradient descent. Perhaps's something similar is going on here with the clustering, where working in batches serves as a benefit in some sense.

2. I found it interesting that Sayak and Ayush mentioned how much more difficult it was to do the data augmentation in Tensorflow (as opposed to PyTorch).

About those referenced papers

"SwaV: Unsupervised Learning of Visual Features by Contrasting Cluster Assignments" here.

"Unsupervised Visual Representation Learning with SwAV" here.


Here's another take on all of this on a slightly different perspective from the gang at Machine Learning Street Talk. Stepping back for a bigger picture.  A very accessible but at the same time surprisingly thoughtful and fascinating discussion.



About those referenced papers

"Do Wide and Deep Networks Learn the Same Things? Uncovering How Neural Network Representations Vary with Width and Depth" here.

"What's in a Loss Function for Image Classification?" here.

"A Simple Framework for Contrastive Learning of Visual Representations" here.

"Big Self-Supervised Models are Strong Semi-Supervised Learners" here.

Comments

Post a Comment

Popular posts from this blog

CycleGAN: a GAN architecture for learning unpaired image to image transformations

Image
 Like the Doublemint Twins touting the joys of Doublemint gum, 2 GANs are surely better than 1 GAN.  Especially if we package them together inside of one meta GAN module.  And this is exactly what the CycleGAN architecture does. Have you ever harbored dark secrets of turning a horse into a zebra?  The CycleGAN was developed to do just that. Learn how to turn a horse into a zebra.  And more. Now right away you can notice a difference between the image to image transformation GAN architectures we've been discussing over the last few posts.  Those last few posts described systems that learn from a database of matched input-output image pairs. And if your goal is to turn an edge representation into a nicely filled in continuous tone image, it's easy to build your database of matched input-output image pairs that your GAN system can then learn off of.  Take a continuous tone image (which will be the output of the database pair entry), then run it through an edge detector algorithm. 
Read more

Pix2Pix: a GAN architecture for image to image transformation

Image
 I thought following up yesterday's TraVelGAN post with a Pix2Pix GAN post would be useful to compare what is going on in the 2 architectures.  Two different approaches to the same problem. I stole this Pix2Pix Overview slide below from an excellent deeplearning.ai GAN course (note that they borrowed it from the original paper) because it gives you a good feel for what is going on inside of the Pix2Pix architecture.    Note how the Generator part is very much like an auto-encoder architecture, but rebuilt using the U-Net architecture features (based on skip-connections) that fastai has been discussing in their courses for several years before it became more widely known to the deep learning community at large  (and which originally came from an obscure medical image segmentation paper) . So the Generator in this Pix2Pix GAN is really pretty sophisticated, consisting of a whole image to image auto-encoder network with U-Net skip connections to generate better image quality at highe
Read more

Smart Fabrics

Image
"What makes smart fabrics revolutionary is that they have the ability to do many things that traditional fabrics cannot, including communicate, transform, conduct energy and even grow."  This is a quote from Rebeccah Pailes-Friedman, who is a professor at Pratt with a research focus on developing wearable technology and smart textiles. Enhancing textiles can be viewed through different lens.  Enhancements can be aesthetic, or performance boosting, or durability boosting. 
Read more