Arthur

[hardmaru @hardmaru]

If you’re interested in learning a bit about Octopuses and their intelligence, check out @pgodfreysmith’s “Other Minds: The Octopus, the Sea, and the Deep Origins of Consciousness”.

I started reading it earlier this year and it is a hard book to put down!
https://t.co/CPKvDNwxwJ

[Sharif Shameem @sharifshameem]

This is mind blowing.

With GPT-3, I built a layout generator where you just describe any layout you want, and it generates the JSX code for you.

W H A T

[Mark Riedl @mark_riedl]

I guess we now shouldn’t expect GPT-3 to be released (though few could even run it in the first place). There will be no external evaluation and scientific progress will be limited.

[Google DeepMind @GoogleDeepMind]

For students and others interested in expanding their knowledge of AI during this period, we thought it might be helpful to ask our researchers what they consider to be the most impactful and insightful resources available to use #AtHomeWithAI (1/9)

[Nick Sharp @nmwsharp]

Polyscope for Python is now live 🎉 Visualize 3D data like meshes and point clouds with just a few lines of code. Now available on pip and conda. https://t.co/TvrhAcHmuJ

[Shane Gu @shaneguML]

Offline RL can help us make actual progress in off-policy RL. This seems like a good dataset for Atari (not-mujoco/bullet) enthusiasts.

[Mauricio Zapata @mezvan]

Gente que aprovecha su conocimiento para hacer cosas curiosas durante la cuarentena

[hardmaru @hardmaru]

Air flow visualization created by @LaVisionInc to demonstrate the effectiveness of using a mask to prevent the spread of cough and sneeze particulates.

“Face masks are reducing COVID-19 infection risk”

https://t.co/EYBv1sBGkw

[Arthur @alpharthur]

@jeffclune Any visiting research/engineering opportunities for current PhD student? Thanks

[Arthur @alpharthur]

Self-Supervised Learning like human creativity.

[hardmaru @hardmaru]

The Illustrated Self-Supervised Learning

Nice overview of recent examples of self-supervised learning with a focus on visual domains (colorization, superres, inpainting, video ordering).

There are endless possibilities of relying less on manual labelling https://t.co/JP3hdv9nUT

[hardmaru @hardmaru]

How to build a brain from scratch

Cool course taught by an experimental psychology dept. Overview of RL, deep learning, generative models, unsupervised learning and parallel topics in learning theory, neuroscience, mental simulation.

Lecture slides/notes https://t.co/tPFI7R6whg

[Arthur @alpharthur]

@hardmaru Because of the nature of the position?

[Arthur @alpharthur]

@reiinakano Congratulations

[KAGAYA @KAGAYA_11949]

ふたご座流星群の流れ星、12連発。
本日未明に撮影した動画から流星が写っている部分をつなげました。
実際のスピードで再生されます。（茨城県にて撮影）

[Arthur @alpharthur]

@kopichan_ backnumberのクリスマスソング🤟🏼

[Arthur @alpharthur]

@enakai00 I am interested

[Arthur @alpharthur]

@mmaciek @Carryveggies Same question too

[Rep. Jim McGovern @RepMcGovern]

RT to tell @SenateMajldr McConnell: bring the bipartisan Hong Kong Human Rights & Democracy Act to the Senate Floor this week! #HongKongProtests #StandWithHongKong

[Emil Wallner @EmilWallner]

Machine learning portfolio tips

1. Good ideas come from ML sources that are a bit quirky.

- NeurIPS from 1987 - 1997
- Stanford’s CS224n & CS231n projects
- Twitter likes from ML outliers
- ML Reddit’s WAYR
- Kaggle Kernels
- Top 15-40% papers on Arxiv Sanity

[Chip Huyen @chipro]

This thread is a combination of 10 free online courses on machine learning that I find the most helpful. They should be taken in order.