Google’s New Imagen AI Outperforms DALL-E on Text-to-Image Generation Benchmarks

Researchers from Google’s Brain Team have announced Imagen, a text-to-image AI model that can generate photorealistic images of a scene given a textual description. Imagen outperforms DALL-E 2 on the COCO benchmark, and unlike many similar models, is pre-… Read more

Similar

Training an AI for the card game Dominion

I'm fascinated by stories in the news of how the world's best AIs for chess and Go were trained by playing against themselves. It seems like a wonderful tool for learning more about the strategy of a game, as well as a useful tool for designing new games. (more…)

Read more »