This article is part of our coverage of the latest inAI research.
It’s free, every week, in your inbox.
You provide DALL-E 2 with a text description, and it generates an image that fits the description.
One of the descriptions ends with as a pencil drawing and the other in photorealistic style.
This kind of consistency shows itself in most examples OpenAI has shared.
Even if the examples on OpenAIs website were cherry-picked, they are impressive.
The results (see the thread below) are fascinating.
But at its heart, it shares the same concept as all otherdeep neural networks: representation learning.
But as has often been seen, deep learning models often learn the wrong representations.
But creating and labeling such a dataset would require immense human effort and is practically impossible.
This is the problem thatContrastive Learning-Image Pre-training(CLIP) solves.
CLIP trains two neural networks in parallel on images and their captions.
DALL-E trains a CLIP model on images and captions.
It then uses the CLIP model to train the diffusion model.
It then tries to generate the image that corresponds to the text.
Since the release ofGPT-2, OpenAI has been reluctant to release its AI models to the public.
GPT-3, its most advanced language model, is only availablethrough an API interface.
Theres no access to the actual code and parameters of the model.
Marcus endorses ahybrid approachthat combines neural networks with symbolic systems.
The DALL-E 2 paper mentions some of the limitations of the model in generating text and complex scenes.
Compositionality is the wall.
The images are beautiful, but no match for the precision of language.https://t.co/uvoXUtETwi
Gary Marcus ?
These demos seem to convince many people that current AI is getting closer and closer to human-level intelligence.
The companys strategic partnership with Microsoft has given it solid channels to monetize some of its technologies, includingGPT-3andCodex.
In ablogpost, Altman suggested a possible DALL-E 2 product launch in the summer.
DALL-E 2 will enable more people to express their creativity without the need for special skills with tools.
it’s possible for you to read the original articlehere.
