Here's a well known example of an "AI" invention (i.e. a novel artifact / discovery found via a search process, genetic search in this case).
@fchollet
-
Deep Learning Models Require Search Processes for Real Novelty
By
–
Mind you, deep learning models are not search processes. They're interpolative — they're big differentiable parametric curves trained with SGD. You need to combine them with a discrete search process (over the right space) to get real novelty.
-
AI Search Processes Generate Novel Discoveries Beyond Human Knowledge
By
–
The fact that a search process over a large enough space could come up with novel (to humans) discoveries wouldn't be contested by anyone familiar with AI, I'd think — the proof of concept came from genetic algorithms in the 1970s. AlphaGo move 37 also comes to mind.
-
Obsessing Over Imaginary X-risks Like Worrying About Aliens
By
–
Imagine being obsessed with imaginary x-risks in this context. It's like worrying about alien invasions in 1960.
-
UX Design Best Practices for Software Libraries
By
–
UX design for software libraries is hard, but there are best practices you can follow. E.g. https://
github.com/keras-team/gov
ernance/blob/master/keras_api_design_guidelines.md
… -
Painfully Accurate Observation Shared by François Chollet
By
–
Painfully accurate https://
x.com/_sn_n/status/1
/_sn_n/status/1657419704615485446
… -
New KerasCV Image Classification Guide with State-of-the-Art Augmentation
By
–
We have a new image classification guide for KerasCV — check it out! Featuring state-of-the-art image augmentation layers. https://
keras.io/guides/keras_c
v/classification_with_keras_cv/
… -
Google Launches New Machine Learning Developer Resources Landing Page
By
–
Google has a brand new landing page for ML developer resources: http://
ai.google/build/machine-
learning
… -
TPUs for efficient language model training
By
–
TPUs are a fast and cost-effective way to train a language model! Check out this post from @huggingface
: