Something from the wired article about @timnitGebru's experience at Google that resonated: expecting Tech to self-regulate on ethical data practices is similar to asking energy companies to self regulate on pollution. The latter is clearly ridiculous, why isn't the former?
— Chris Holdgraf (@choldgraf) June 12, 2021
MuZero removed simulators in MBRL vs AlphaGo. VQ Models for Planning generalize to partial observable & stochastic environments. How?
— Oriol Vinyals (@OriolVinyalsML) June 11, 2021
1. Discretize states w/ VQVAE
2. Train a LM over states
3. Plan w/ MCTS using the LM
Led by @yazhe_li & @sherjilozair https://t.co/thvB6Ke1EA pic.twitter.com/tsXGcrweTZ
Today, we’re introducing TextStyleBrush, the first self-supervised AI model that replaces text in existing images of both scenes and handwriting — in one shot — using just a single example word: https://t.co/0QfLraAQvV pic.twitter.com/FNDJxNC20S
— Facebook AI (@facebookai) June 11, 2021
Transformed CNNs: recasting pre-trained convolutional layers with self-attention
— AK (@ak92501) June 11, 2021
pdf: https://t.co/NNk9Ngp90o
abs: https://t.co/RM4nyVA3N6
+2.2% top-1 on ImageNet-1k for a ResNet50-RS as well as substantially improved robustness +11% top-1 on ImageNet-C pic.twitter.com/0z6GRrgjC7
MusicBERT: Symbolic Music Understanding with Large-Scale Pre-Training
— AK (@ak92501) June 11, 2021
pdf: https://t.co/nvO4RUj540
abs: https://t.co/vjEeRlX9Ud
a largescale pre-trained model for symbolic music understanding, sota performance on four evaluated symbolic music understanding tasks pic.twitter.com/cF7I2vwjx0
Map shows US slave population in 1790 and 1860. A truly dark chapter in US history. Source: https://t.co/R2H7pX6qPd pic.twitter.com/PG9PwnSBEt
— Simon Kuestenmacher (@simongerman600) June 10, 2021
When MCMC fails: The advice we’re giving is wrong. Here’s what we you should be doing instead. (Hint: it’s all about the folk theorem.) https://t.co/2juL4UxJV2
— Andrew Gelman et al. (@StatModeling) June 10, 2021
The development of hundreds AI/ML/Stats algorithms for diagnosing and prognosticating COVID-19 patients has been a disaster (with very, very few exceptions). Hopefully the worst example of research waste I will ever see pic.twitter.com/IFRew5OCZb
— Maarten van Smeden (@MaartenvSmeden) June 10, 2021
AdaMatch: A Unified Approach to Semi-Supervised
— AK (@ak92501) June 10, 2021
Learning and Domain Adaptation
pdf: https://t.co/9BUJQK3SdZ
abs: https://t.co/G1AOXAopye
a general method designed to boost accuracy on domain shifts when given access to unlabeled data from the new domain pic.twitter.com/n8PCfST3ql
Happy to introduce CoAtNet: combining convolution and self-attention in a principled way to obtain better capacity and better generalization.
— Mingxing Tan (@tanmingxing) June 10, 2021
88.56% top-1 with ImageNet21K (13M imgs), matching ViT-huge with JFT (300M imgs).
Paper: https://t.co/AQE33LuzSr pic.twitter.com/YEly0cSaTp
Knowledge distillation: A good teacher is patient and consistent
— AK (@ak92501) June 10, 2021
pdf: https://t.co/5PsmI4NGFc
abs: https://t.co/5AkD5om0wH
sota ResNet-50 model for ImageNet, which achieves 82.8% top-1 accuracy pic.twitter.com/aZLj8gVThJ
A comprehensive overview of Transformer variants.https://t.co/VXkshkSorT#AI #ML #DL #DataScience pic.twitter.com/1I11qeWr2q
— Bojan Tunguz (@tunguz) June 9, 2021
