Company Filing History:
Years Active: 2022-2025
Title: Innovations of Junnan Li: A Pioneer in Image Classification Technology
Introduction
Junnan Li, an accomplished inventor based in Singapore, has made significant contributions to the field of image classification technology, evidenced by his impressive portfolio of 14 patents. His work focuses primarily on unsupervised and semi-supervised learning methodologies that leverage innovative techniques for improving image classification systems.
Latest Patents
Among Junnan's most notable recent inventions are two pioneering patents. The first, titled "Systems and methods for masked self-training of unsupervised image classification," introduces a masked self-training (MaST) approach. This method harnesses two complementary sources of supervision: pseudo-labels and raw image pixels. By optimizing three objectives, MaST fine-tunes a pre-trained classification model on unlabeled images, effectively enhancing the model's predictive capabilities.
The second patent, "Systems and methods for semi-supervised learning with contrastive graph regularization," presents a 'Co-training' mechanism that allows for the joint learning of two representations of training data. This method generates class probabilities and low-dimensional embeddings, utilizing memory-smoothed pseudo-labels and contrastive learning to improve the quality of image representations.
Career Highlights
Throughout his career, Junnan has worked with some prestigious organizations. Notably, his tenure at Salesforce, Inc. has been marked by advancements in image classification technology. His contributions have also been recognized and utilized by various teams in the industry, further solidifying his reputation as an innovative problem-solver.
Collaborations
Collaborating with talented individuals such as Chu Hong Hoi and Dongxu Li, Junnan has enhanced his research output and introduced transformative ideas within his projects. These partnerships have fostered an environment of creativity and collaboration, allowing for the exploration of novel concepts in machine learning and image recognition.
Conclusion
In summary, Junnan Li stands out as a leading inventor in the realm of image classification technology, with a strong portfolio of patents that reflect his innovative spirit and expertise. His latest breakthroughs are paving the way for more efficient and accurate image classification systems, showcasing the impact of his work on both academia and industry.
Inventor’s Patent Attorneys refers to legal professionals with specialized expertise in representing inventors throughout the patent process. These attorneys assist inventors in navigating the complexities of patent law, including filing patent applications, conducting patent searches, and protecting intellectual property rights. They play a crucial role in helping inventors secure patents for their innovative creations.