Jongseong Jang

Title: Jongseong Jang: Innovator in Online Class-Incremental Continual Learning

Introduction

Jongseong Jang is a notable inventor based in Hanam-si, South Korea. He has made significant contributions to the field of machine learning, particularly in the area of online class-incremental continual learning. His innovative approach addresses the challenges of preserving knowledge while adapting to new data.

Latest Patents

Jang holds a patent for a method titled "Online class-incremental continual learning with adversarial Shapley value." This method involves scoring training data samples to maintain latent decision boundaries for previously observed classes while facilitating learning from new images in an online data stream. The process includes receiving new input batches, performing memory retrieval to retain previously learned knowledge, and updating memory by selecting a small set of data for retention. Additionally, the method employs data valuation based on KNN-SV for strategic data selection.

Career Highlights

Throughout his career, Jongseong Jang has worked with prominent organizations, including LG Electronics Inc. and the University of Toronto. His experience in these institutions has allowed him to refine his skills and contribute to cutting-edge research in machine learning.

Collaborations

Jang has collaborated with talented individuals such as Dongsub Shim and Zheda Mai, enhancing his work through shared expertise and innovative ideas.

Conclusion

Jongseong Jang is a pioneering inventor whose work in online class-incremental continual learning is shaping the future of machine learning. His contributions are vital for advancing the field and addressing the complexities of data management in real-time applications.