This inventor holds 1 USPTO granted patent. Top assignee: Meta Platforms, Inc.. Active years: 2025.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2025
Title: Innovations of Yanghao Li in Computer Vision
Introduction
Yanghao Li is a prominent inventor based in Sunnyvale, CA. He has made significant contributions to the field of computer vision, particularly through his innovative patent. His work is recognized for its potential to enhance machine learning models.
Latest Patents
Yanghao Li holds a patent titled "Masked Autoencoders for Computer Vision." This invention involves a computing system that accesses a plurality of images for pre-training a machine-learning model. The model includes an encoder and a decoder. During the pre-training process, the system divides each image into a set of patches. It selects a first subset of patches to be visible and a second subset to be masked. The encoder processes the visible patches to generate corresponding latent representations. The decoder then reconstructs the masked patches using these representations and mask tokens. This innovative approach allows for the generation of reconstructed images, which are compared to the original images to update the model.
Career Highlights
Yanghao Li is currently employed at Meta Platforms, Inc. His work at this leading technology company has allowed him to explore and develop cutting-edge technologies in machine learning and computer vision. His contributions are vital to advancing the capabilities of artificial intelligence.
Collaborations
Yanghao has collaborated with notable colleagues in the field, including Xinlei Chen and Kaiming He. These collaborations have fostered an environment of innovation and creativity, leading to significant advancements in their respective areas of expertise.
Conclusion
Yanghao Li's contributions to computer vision through his patent on masked autoencoders exemplify the innovative spirit of modern technology. His work at Meta Platforms, Inc. continues to push the boundaries of what is possible in machine learning.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
