Linnan Wang

Title: Linnan Wang: Innovator in Neural Network Training

Introduction

Linnan Wang is a prominent inventor based in Piscataway, NJ (US). He has made significant contributions to the field of artificial intelligence, particularly in the training of deep neural networks. His innovative approach has garnered attention in the tech community.

Latest Patents

Linnan Wang holds a patent titled "Accelerating deep neural network training with inconsistent stochastic gradient descent." This patent describes techniques for training a convolutional neural network using an inconsistent stochastic gradient descent (ISGD) algorithm. The ISGD algorithm dynamically adjusts the training effort for batches based on a determined loss, classifying them into two sub-states: well-trained or under-trained. This method allows for more iterations on under-trained batches while reducing iterations for well-trained ones, enhancing the overall efficiency of the training process.

Career Highlights

Linnan Wang is associated with NEC Corporation, where he applies his expertise in neural network training. His work has been instrumental in advancing the capabilities of machine learning technologies. He has a strong focus on improving the performance and efficiency of deep learning models.

Collaborations

Linnan has collaborated with notable colleagues, including Yi Yang and Renqiang Min. Their combined efforts contribute to the innovative projects at NEC Corporation.

Conclusion

Linnan Wang's contributions to the field of deep learning and neural network training are noteworthy. His patent on the ISGD algorithm exemplifies his commitment to advancing technology in artificial intelligence. His work continues to influence the development of more efficient machine learning techniques.