Formal Reasoning and Program Synthesis

 This is a discussion between Christian Szegedy of Google Research and the gang at Machine Learning Street Talk.  It covers a lot of ground. 

One area is whether we can use neural net systems to auto-formalize the methodology of practicing and utilizing mathematics.  Wolfram research is jumping up and down at this point, saying 'hey guys, we already did that' (the auto-formalize part).

Another area is discussion about Transformers and how they work.  Hot topic. Perhaps over-hyped in my humble opinion (ai lemmings running towards a transformer cliff), but that's for another post.

I'm not sure what i think about this notion of neural networks 'running programs'. Tim's comments at the very beginning about neural networks being systems that interpolate on a lower dimensional manifold get's to the exact heart of how they work.  If someone writes a program to implement a radial basis function interpolator, do we really care about the program, or just the function it is interpolating?  

But we all need to keep an open mind, and there are lot's of interesting things to learn listening to this discussion.

Certainly the notion that we have been programming computers essentially the dame way since the 1950's is pretty spot on. Please give us better ways to do this task.



Observations

1:  Another reason why we have a special holiday every year at HTC that celebrates Jurgen Schmidhuber and his various contributions to the field.

2: Ok, reasoning in latent embedded space. I need to think about that more.

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