AI Times has published Yongduck Kim's patent analysis of OpenAI's technology for performing automated tasks using video.
The analysis concerns training AI to perform interface tasks using unlabeled video. OpenAI's U.S. Patent No. 12,315,255 B2 describes video-based learning using pseudo-labeling and an inverse dynamics model (IDM).
Yongduck Kim patent attorney explained the technical structure of the patent easily, and introduced the possibilities for use in various fields such as web automation, game interface, and robotics control.
You can find the full article at the link below.
Read the column
This article reflects the information available when it was published. Contact us to discuss your circumstances.