davidADSP's profile picture. Author of Generative Deep Learning: Teaching Machines how to Paint, Write, Compose and Play (O'Reilly), #generativeAI, 
Founding Partner of ADSP.

David Foster

@davidADSP

Author of Generative Deep Learning: Teaching Machines how to Paint, Write, Compose and Play (O'Reilly), #generativeAI, Founding Partner of ADSP.

Spot the data viz fail 🤦‍♂️@BBC @BBCPolitics @BBCNews

davidADSP's tweet image. Spot the data viz fail 🤦‍♂️@BBC @BBCPolitics @BBCNews

Kicking off 2023 with more exciting generative publications! Here's MUSE (text-to-image) from Google muse-model.github.io No diffusion models - just VQ-GAN tokenisers and pure Transformers all the way! #generativeAI, #AI #machinelearning #deeplearning #DALLE #ChatGPT


Updated version of the Generative AI timeline, featuring more LLMs 💪 #GenerativeAI #AI #DeepLearning #LLM #ChatGPT #gpt4 #GPT3 #stablediffusion #dalle

davidADSP's tweet image. Updated version of the Generative AI timeline, featuring more LLMs 💪

#GenerativeAI #AI #DeepLearning #LLM #ChatGPT #gpt4 #GPT3 #stablediffusion #dalle

Love this - this blend of #generative #AI, #RL and active inference will be a fascinating area of development over the next 2-3 years.

GPT3 is powerful but blind. The future of Foundation Models will be embodied agents that proactively take actions, endlessly explore the world, and continuously self-improve. What does it take? In our NeurIPS Outstanding Paper “MineDojo”, we provide a blueprint for this future:🧵



David Foster reposted

Summary and visualization of different normalization techniques: batch norm, layer norm, instance norm and group norm; mean and variance is calculated among blocks of the same color.

xucao42's tweet image. Summary and visualization of different normalization techniques: batch norm, layer norm, instance norm and group norm; mean and variance is calculated among blocks of the same color.

A bright cold day in April, and the clocks were striking thirteen, digital art #dalle2

davidADSP's tweet image. A bright cold day in April, and the clocks were striking thirteen, digital art #dalle2

Love this. Diffusion models are truly beautiful and the future of generative modelling - this is a fantastic tutorial to learn all about them!

New tutorial on keras.io: generating images using denoising diffusion implicit models. Start experimenting with your own image generation models! keras.io/examples/gener…



David Foster reposted

Unmodernizing a Jackson Pollock painting with #dalle text diffs


Cutting-edge just got sharper #dalle #generativedeeplearning @openai

DALL·E 2 — generate any image from a text description. Imagination is the limit. "A Shiba Inu dog wearing a beret and black turtleneck" "A photo of a quaint flower shop storefront with a pastel green and clean white facade and open door and big window" openai.com/dall-e-2/

gdb's tweet image. DALL·E 2 — generate any image from a text description. Imagination is the limit.

"A Shiba Inu dog wearing a beret and black turtleneck"
"A photo of a quaint flower shop storefront with a pastel green and clean white facade and open door and big window"

openai.com/dall-e-2/
gdb's tweet image. DALL·E 2 — generate any image from a text description. Imagination is the limit.

"A Shiba Inu dog wearing a beret and black turtleneck"
"A photo of a quaint flower shop storefront with a pastel green and clean white facade and open door and big window"

openai.com/dall-e-2/


hey @GMJanGustafsson - I'm trying to buy my brother a mug with your face on it for Christmas, but it won't let me ship to the UK through the @chess24com site. Do you have any spares at home that I could purchase off you? Hope they haven't sold out already 😬 #chess #xmasmiracle

davidADSP's tweet image. hey @GMJanGustafsson - I'm trying to buy my brother a mug with your face on it for Christmas, but it won't let me ship to the UK through the @chess24com site.

Do you have any spares at home that I could purchase off you? Hope they haven't sold out already 😬

#chess #xmasmiracle

Loading...

Something went wrong.


Something went wrong.