Duanduan Yang

Title: Duanduan Yang: Innovator in Neural Networks and Convolution Calculations

Introduction

Duanduan Yang is a prominent inventor based in San Jose, CA, known for his contributions to the fields of neural networks and convolution calculations. With a total of 6 patents to his name, Yang has made significant strides in enhancing the efficiency and effectiveness of computational methods in technology.

Latest Patents

Yang's latest patents include a "System and method for hashed compressed weighting matrix in neural networks." This innovative method involves receiving an input from a vector of inputs, determining a table index based on the input, and retrieving a hash table from a plurality of hash tables. The process continues with determining an entry index of the hash table based on an index matrix, which includes one or more index values corresponding to a vector in the hash table. Additionally, Yang has developed a "System and method for cached convolution calculation." This method identifies input words associated with a user query and retrieves pre-computed element-wise products from a cache for high-frequency words, thereby optimizing the convolution operation.

Career Highlights

Throughout his career, Duanduan Yang has worked with notable companies such as Konica Minolta Laboratory U.S.A., Inc. and Samsung Electronics Co., Ltd. His experience in these organizations has allowed him to refine his skills and contribute to groundbreaking innovations in technology.

Collaborations

Yang has collaborated with talented individuals in his field, including Saman Sarraf and Xiang Sun. These partnerships have fostered an environment of creativity and innovation, leading to the development of advanced technologies.

Conclusion

Duanduan Yang's work in neural networks and convolution calculations exemplifies the impact of innovative thinking in technology. His patents and collaborations highlight his commitment to advancing computational methods, making him a significant figure in the field.