Company Filing History:
Years Active: 2019
Title: Xuesong Yang: Innovator in Language Understanding and Dialogue Management
Introduction
Xuesong Yang is a notable inventor based in Urbana, IL (US). He has made significant contributions to the field of natural language processing, particularly in the area of dialogue management systems. His innovative approach combines advanced machine learning techniques to enhance user interaction with technology.
Latest Patents
Xuesong Yang holds a patent titled "Joint language understanding and dialogue management using binary classification based on forward and backward recurrent neural network." This patent describes a processing unit capable of operating an end-to-end recurrent neural network (RNN) with limited contextual dialog memory. The system can be jointly trained using supervised signals for user slot tagging, intent prediction, and system action prediction. The advantages of this joint model over separate models include improved natural language understanding (NLU) and dialogue management, as it captures expressive feature representations that mitigate the effects of noisy output from NLU. By back-propagating errors associated with system action prediction to the NLU model, the joint model effectively predicts user intent through binary classification.
Career Highlights
Xuesong Yang is currently employed at Microsoft Technology Licensing, LLC, where he continues to develop innovative solutions in language understanding. His work has been instrumental in advancing the capabilities of dialogue systems, making them more efficient and user-friendly.
Collaborations
Xuesong has collaborated with notable colleagues, including Xiujun Li and Paul Anthony Crook. Their combined expertise has contributed to the success of various projects within the realm of natural language processing.
Conclusion
Xuesong Yang's contributions to language understanding and dialogue management exemplify the impact of innovative thinking in technology. His work continues to shape the future of human-computer interaction, making it more intuitive and effective.