Company Filing History:
1 out of 5,004 
1 out of 5 Years Active: 2020
Title: Inventor Profile: Junlong Kang
Introduction
Junlong Kang is an innovative inventor based in Beijing, China, known for his contributions to the field of artificial intelligence, particularly in the development of hardware accelerators for neural networks. With a total of two patents to his name, Kang has made significant advancements in the implementation and acceleration of recurrent neural networks (RNN) and Long Short Term Memory (LSTM) networks.
Latest Patents
Kang's latest patents include groundbreaking inventions aimed at enhancing the performance of neural networks. The first patent, titled "Hardware Accelerator for Compressed RNN on FPGA," focuses on an overall design processing method that involves matrix decoding, matrix-vector multiplication, vector accumulation, and activation functions. This invention presents a comprehensive approach to implementing and accelerating RNNs by leveraging embedded FPGAs.
The second patent, "Hardware Accelerator for Compressed LSTM," outlines an innovative accelerator that includes a sparse matrix-vector multiplication module to perform essential operations within LSTMs. This accelerator not only accumulates results through an addition tree module but also implements a non-linear operation module to pass the accumulated results through an activation function. The design adopts a pipeline approach, ensuring efficient overlap of data transfer and computation for compressed LSTMs.
Career Highlights
Junlong Kang has built a notable career in technology, with significant roles at leading companies in the industry. He has worked with Xilinx, Inc., where he contributed to the advancement of FPGA technology. Additionally, he has been associated with Beijing Deephi Intelligent Technology Co., Ltd., further expanding his expertise in intelligent systems and hardware accelerators.
Collaborations
Throughout his career, Kang has collaborated with several prominent professionals in his field. Notable coworkers include Song Han and Yi Shan, whose collective efforts in research have contributed to the success and innovation of their projects.
Conclusion
Junlong Kang's work in creating hardware accelerators for RNNs and LSTMs exemplifies his commitment to advancing artificial intelligence technology. With his patents and experiences, Kang continues to be a significant figure in the landscape of innovations in neural network acceleration. His work not only enhances computational efficiency but also paves the way for future developments in intelligent systems and artificial intelligence applications.
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.