How policies designed to improve academia are actually messing it up liebertpub.com/doi/full/10.10… "If a critical mass of scientists become untrustworthy, a tipping point is possible in which the scientific enterprise itself becomes inherently corrupt and public trust is lost"

Announcing PyTorch Hub! Want ResNet, ResNext, BERT, GPT, PGAN, Tacotron, DenseNet, MobileNet...? - Pull models with 1 line of code. - Publish your models by sending a pull request. More info here:... pytorch.org/blog/towards-r…

Presenting our work on Tuesday, August 6th at KDD conference on 'Cloud-Based Data Science at the Speed of Thought Using RAPIDS - the Open GPU Data Science Ecosystem'. Do join our tutorial Microsoft @AzureML #kdd #sigkdd #anchorage #kdd2019 #azureml

Published some of our distributed training performance improvements using SRIOV on AzureML. Feel free to ask questions, if any. techcommunity.microsoft.com/t5/Azure-AI/Ac… @AzureML Microsoft

Microsoft researchers and engineers release Zero Redundancy Optimizer (ZeRO) and DeepSpeed library, a system able to train 100-billion-parameter deep learning models. Learn about this breakthrough and how it led to Turing Natural Language Generation: aka.ms/AA79s5c

A couple papers from 2016. Sort of like those doomsday movies with the ignored scientist warning of impending doom.

Today, we announced the Abnormal Integrated Cloud Email Security (ICES) platform, the only cloud-native solution for comprehensively and precisely stopping all email threats. Learn more in the Abnormal Blog: bit.ly/ICESLaunch

Home after a year! <3 :D — traveling to New Delhi, India from San Francisco International Airport (SFO) fb.me/1UjwcVIUk

Montreal AI pioneer Yoshua Bengio is launching deep learning incubator wired.com/2016/10/ai-pio…

Turning big data into manageable data: on.forbes.com/60148MwYp

DARN: Deep Adversarial ResNet for Intrinsic Image Decomposition arxiv.org/abs/1612.07899 creativeai.net/posts/AK7pngds… NN splits albedo+shading!

The video of my keynote talk at NIPS 2016 is now available. Title: predictive learning. Abstract:... fb.me/5TAG4kCiM

To all Researchers: If US Muslim ban remains, lets move our conferences elsewhere. Science must be open to all.

Regarding the meeting at the White House:

One way to think about ML applications: the cost of a single mistake Ads: <1 cent Drilling for oil: $5m-$100m Medical: $1k-a human life

stanforddaily.com/2017/02/23/a-s…

Jennifer Widom and I had taught two of first MOOCs in 2011. Great to see her as new Stanford Eng Dean! news.stanford.edu/2017/02/27/com…

The AI Index is out! Page after page of interesting charts showing AI trends. aiindex.org For example, since 2013, the share of US jobs requiring AI skills has grown 4.5x! Yoav Shoham