New paper analyzing sample-based metrics for evaluating generative models, from the Cornell group. Tests if they can detect things like overfitting and mode collapse. Should be required reading for everyone working on generative models. arxiv.org/abs/1806.07755

Check out blog.openai.com/glow/, my work with Prafulla Dhariwal on improving flow-based generative models with invertible 1x1 convolutions. youtu.be/exJZOC3ZceA

Our latest work on ‘Measuring abstract reasoning in neural networks’ has just been published at #icml2018. As always, it was a privilege to collaborate with Adam Santoro, Felix Hill, Ari Morcos and Tim Lillicrap. Paper: tinyurl.com/y7qfd2d7 Blog post: deepmind.com/blog/measuring…

In 1951, Bertrand Russel took to the The New York Times to argue that the best answer to fanaticism was a calm search for truth. His Ten Commandments of Liberal Inquiry could not be more relevant today. (Number 6 will blow your mind! ;) ) Thread.

i've never seen a clearer explanation of plato's cave

Today we are excited to release video recordings of lectures from "Advanced Deep Learning and Reinforcement Learning", a course on deep RL taught at @UCL earlier this year by DeepMind researchers: youtube.com/playlist?list=… Enjoy!

A visual exploration of Gaussian Processes: beautiful interactive plots and a brief tutorial to make GPs more approachable jgoertler.com/visual-explora…

Took a while (don't ask) but here they are: Notes from "Science of Deep Learning" class co-taught with Constantinos Daskalakis now available: people.csail.mit.edu/madry/6.883/. More coming soon (promise!). Feedback very welcome! Thanks to Andrew Ilyas for heroic effort on doing final revisions.

Paper "What is it like to be a bat?" Thomas Nagel's thought experiment about consciousness is still as relevant today as it was when it was first published in 1974. fermatslibrary.com/s/what-is-it-l…

Today at 11:30 EST / 16:30 GMT we'll be presenting our poster about our work “On Calibration and Out-of-domain Generalization” at #NeurIPS2021, come visit! neurips.cc/virtual/2021/p… Yoav Wald Amir Feder Daniel Greenfeld

Best illusion I've seen, maybe ever. The blue and black lines are NOT moving. Insanely good.

Encoder-decoder GANs architectures still don't fix the theoretical problems in GANs framework such as mode collapse. Encoders may produce nonsense codes and the discriminator is none the wiser. Blog post offconvex.org/2018/03/12/big… and ICLR'18 paper openreview.net/forum?id=BJehN…

GESTALT PRINCIPLES THREAD! Gestalt is the idea that we see the whole of something before the individual parts. PROXIMITY (1/8) When objects are close to each other, they tend to be perceived together in a group. Use white space to separate groups. Reduce it to group elements.

Nesting probabilistic programs allows us to model agents reasoning about other agents, but current inference engines typically give invalid estimates. Check out how to do things correctly in my new paper arxiv.org/abs/1803.06328

Check out Adversarial Logit Pairing, the new state of the art defense against adversarial examples on ImageNet, by Harini Kannan Alexey Kurakin and I: arxiv.org/abs/1803.06373

Simple GAN inversion experiments easily show that ***all*** real images (except for zero measure subset) have 0 probability of being generated by a GAN (off the manifold). What does this say about the promise (or lack thereof) of training models based on GAN-generated datasets.

2nd thread on evaluating GAN papers (1st thread hit max thread length)

By learning to write programs that generate images our artificial agents can reason about how digits, characters and portraits are constructed. Read the blog: deepmind.com/blog/learning-…