AssemblyAI
DALL-E 2 has arrived in the AI world with a bang. It is one of the best generative models we have seen to date. But how does this magical model work? In this video, we will take a look into the architecture of DALL-E 2, to understand the working principles.
00:00 Overview
00:34 What can DALL-E 2 do?
00:55 Architecture overview
01:27 CLIP embeddings
03:05 The prior
04:24 Why do we need the prior?
05:20 The decoder
06:13 How are variations created?
06:56 Model evaluation
07:36 Limitations and risks of DALL-E 2
09:21 Benefits of DALL-E 2
10:00 A question for you!
Example images are from DALL-E 2 announcement blog post (https://openai.com/dall-e-2/) and the DALL-E 2 paper (https://cdn.openai.com/papers/dall-e-2.pdf)
The analysis of DALL-E 2 limitations and risks – https://github.com/openai/dalle-2-preview/blob/main/system-card.md
Would you like to read this information instead? Check out the blog post on DALL-E 2 👇
https://www.assemblyai.com/blog/how-dall-e-2-actually-works/
Get your Free Token for AssemblyAI Speech-To-Text API 👇https://www.assemblyai.com/?utm_source=youtube&utm_medium=referral&utm_campaign=yt_mis_28
▬▬▬▬▬▬▬▬▬▬▬▬ CONNECT ▬▬▬▬▬▬▬▬▬▬▬▬
🖥️ Website: https://www.assemblyai.com
🐦 Twitter: https://twitter.com/AssemblyAI
🦾 Discord: https://discord.gg/Cd8MyVJAXd
▶️ Subscribe: https://www.youtube.com/c/AssemblyAI?sub_confirmation=1
🔥 We’re hiring! Check our open roles: https://www.assemblyai.com/careers
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
#MachineLearning #DeepLearning