Skip to content

Top New Best Ask Show Jobs

MrUssek

208 karmaJoined April 1, 201924 submissions

MrUssek | Better HN

Recent submissions

1

China has trained a 10 trillion parameter language model (opens in new tab)

(twitter.com)

4MrUssek4y ago0

2

What is your backup if the tech industry crashes?

4MrUssek4y ago10

3

The Future of Deep Learning Is Photonic (opens in new tab)

(spectrum.ieee.org)

1MrUssek4y ago0

4

Separating MNIST digits using Optimal Transport (opens in new tab)

(mrussek.com)

1MrUssek4y ago0

5

Enigma: GPT-2 trained on 10K Nature Papers: Can you spot the difference? (opens in new tab)

(stefanzukin.com)

183MrUssek4y ago105

6

GShard: Scaling giant models with conditional computation and automatic sharding (opens in new tab)

(arxiv.org)

112MrUssek5y ago35