Загрузка страницы

[ML News] Microsoft trains 530B model | ConvMixer model fits into single tweet | DeepMind profitable

#mlnews #turingnlg #convmixer

Your latest upates on what's happening in the Machine Learning world.

OUTLINE:
0:00 - Intro
0:16 - Weights & Biases raises on 1B valuation (sponsored)
2:30 - Microsoft trains 530 billion parameter model
5:15 - StyleGAN v3 released
6:45 - A few more examples may be worth billions of parameters
8:30 - ConvMixer fits into a tweet
9:45 - Improved VQGAN
11:25 - William Shatner AI chats about his life
12:35 - Google AI pushes material science
14:10 - Gretel AI raises 50M for privacy protection
16:05 - DeepMind's push into ML for biology
19:00 - Schmidhuber laudates Kunihiko Fukushima for Bower Award
21:30 - Helpful Things
22:25 - Mosaic ML out of stealth mode
23:55 - First German self-driving train
24:45 - Ex-Pentagon Chief: China has already won
26:25 - DeepMind becomes profitable

Sponsor: Weights & Biases
https://wandb.com

References:
Microsoft Trains 530B Parameter Model
https://www.microsoft.com/en-us/research/blog/using-deepspeed-and-megatron-to-train-megatron-turing-nlg-530b-the-worlds-largest-and-most-powerful-generative-language-model/

StyleGAN 3 Code Released
https://nvlabs.github.io/stylegan3/
https://github.com/NVlabs/stylegan3
https://colab.research.google.com/github/ouhenio/StyleGAN3-CLIP-notebook/blob/main/StyleGAN3%2BCLIP.ipynb#scrollTo=V_rq-N2m0Tlb

When do labels help?
https://arxiv.org/pdf/2110.04374.pdf

ml_paper.bruh
https://openreview.net/pdf?id=TVHS5Y4dNvM

Improved VQGAN
https://openreview.net/pdf?id=pfNyExj7z2

William Shatner "AI" & Storyfile
https://www.livescience.com/william-shatner-ai-chat?fbclid=IwAR19yapmIotCTL9NIpz1xy2Ayq3H869i7TU34Vm-obxRaCLeX5YMDR_Wl-Y&utm_source=pocket_mylist
https://www.storyfile.com/

GoogleAI Finds Complex Metal Oxides
https://ai.googleblog.com/2021/10/finding-complex-metal-oxides-for.html

GretelAI raises 50M Series B
https://techcrunch.com/2021/10/07/gretel-ai-raises-50m-for-a-platform-that-lets-engineers-build-and-use-synthetic-datasets-to-ensure-the-privacy-of-their-actual-data/
https://gretel.ai/
https://gretel.ai/blog/why-privacy-by-design-matters-more-than-ever

DeepMind's Push in ML for Bio
https://www.biorxiv.org/content/10.1101/2021.10.04.463034v1
https://deepmind.com/blog/article/enformer

Kunihiko Fukushima wins Bower Award: Schmidhuber Congratulates
https://www.fi.edu/laureates/kunihiko-fukushima
https://www.youtube.com/watch?v=ysOw6lNWx2o

Helpful Things
https://github.com/UKPLab/beir#beers-features
https://arxiv.org/pdf/2104.08663.pdf
https://bayesoptbook.com/
https://github.com/nvlabs/imaginaire/
https://github.com/NVlabs/imaginaire/blob/master/projects/gancraft/README.md

MosaicML out of Stealth Mode
https://www.mosaicml.com/
https://www.mosaicml.com/blog/founders-blog
https://app.mosaicml.com/library/imagenet
https://github.com/mosaicml/composer
https://mosaicml-composer.readthedocs-hosted.com/en/stable/

Germany's first self-driving train
https://techxplore.com/news/2021-10-germany-unveils-self-driving.html

Ex-Pentagon Chief: China has already won tech war
https://nypost.com/2021/10/11/pentagon-software-chief-nicolas-chaillan-resigns/

DeepMind becomes profitable
https://bdtechtalks.com/2021/10/07/google-deepmind-2020-earnings/

Links:
TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
Discord: https://discord.gg/4H8xxDF
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.minds.com/ykilcher
Parler: https://parler.com/profile/YannicKilcher
LinkedIn: https://www.linkedin.com/in/ykilcher
BiliBili: https://space.bilibili.com/1824646584

If you want to support me, the best thing to do is to share out the content :)

If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this):
SubscribeStar: https://www.subscribestar.com/yannickilcher
Patreon: https://www.patreon.com/yannickilcher
Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq
Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2
Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m
Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n

Видео [ML News] Microsoft trains 530B model | ConvMixer model fits into single tweet | DeepMind profitable канала Yannic Kilcher
Показать
Комментарии отсутствуют
Введите заголовок:

Введите адрес ссылки:

Введите адрес видео с YouTube:

Зарегистрируйтесь или войдите с
Информация о видео
20 октября 2021 г. 18:40:13
00:27:51
Другие видео канала
WHO ARE YOU? 10k Subscribers Special (w/ Channel Analytics)WHO ARE YOU? 10k Subscribers Special (w/ Channel Analytics)Datasets for Data-Driven Reinforcement LearningDatasets for Data-Driven Reinforcement LearningReinforcement Learning with Augmented Data (Paper Explained)Reinforcement Learning with Augmented Data (Paper Explained)The Odds are Odd: A Statistical Test for Detecting Adversarial ExamplesThe Odds are Odd: A Statistical Test for Detecting Adversarial ExamplesExpire-Span: Not All Memories are Created Equal: Learning to Forget by Expiring (Paper Explained)Expire-Span: Not All Memories are Created Equal: Learning to Forget by Expiring (Paper Explained)REALM: Retrieval-Augmented Language Model Pre-Training (Paper Explained)REALM: Retrieval-Augmented Language Model Pre-Training (Paper Explained)Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation (Paper Explained)Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation (Paper Explained)[Classic] Playing Atari with Deep Reinforcement Learning (Paper Explained)[Classic] Playing Atari with Deep Reinforcement Learning (Paper Explained)Symbolic Knowledge Distillation: from General Language Models to Commonsense Models (Explained)Symbolic Knowledge Distillation: from General Language Models to Commonsense Models (Explained)Gradient Origin Networks (Paper Explained w/ Live Coding)Gradient Origin Networks (Paper Explained w/ Live Coding)Perceiver: General Perception with Iterative Attention (Google DeepMind Research Paper Explained)Perceiver: General Perception with Iterative Attention (Google DeepMind Research Paper Explained)PonderNet: Learning to Ponder (Machine Learning Research Paper Explained)PonderNet: Learning to Ponder (Machine Learning Research Paper Explained)ALiBi - Train Short, Test Long: Attention with linear biases enables input length extrapolationALiBi - Train Short, Test Long: Attention with linear biases enables input length extrapolationListening to You! - Channel Update (Author Interviews)Listening to You! - Channel Update (Author Interviews)On the Measure of Intelligence by François Chollet - Part 1: Foundations (Paper Explained)On the Measure of Intelligence by François Chollet - Part 1: Foundations (Paper Explained)[ML News] Uber: Deep Learning for ETA | MuZero Video Compression  | Block-NeRF | EfficientNet-X[ML News] Uber: Deep Learning for ETA | MuZero Video Compression | Block-NeRF | EfficientNet-XGrowing Neural Cellular AutomataGrowing Neural Cellular Automata[ML News] DeepMind's Flamingo Image-Text model | Locked-Image Tuning | Jurassic X & MRKL[ML News] DeepMind's Flamingo Image-Text model | Locked-Image Tuning | Jurassic X & MRKLAvoiding Catastrophe: Active Dendrites Enable Multi-Task Learning in Dynamic Environments (Review)Avoiding Catastrophe: Active Dendrites Enable Multi-Task Learning in Dynamic Environments (Review)AMP: Adversarial Motion Priors for Stylized Physics-Based Character Control (Paper Explained)AMP: Adversarial Motion Priors for Stylized Physics-Based Character Control (Paper Explained)SupSup: Supermasks in Superposition (Paper Explained)SupSup: Supermasks in Superposition (Paper Explained)
Яндекс.Метрика