Imai Haruki

Title: Imai Haruki: Innovator in Neural Network Training

Introduction

Imai Haruki is a prominent inventor based in Yokohama, Japan. He has made significant contributions to the field of artificial intelligence, particularly in the efficient training of neural networks. His innovative approach has garnered attention in the tech community.

Latest Patents

Imai Haruki holds a patent for an invention titled "Efficient parallel training of a network model on multiple graphics processing units." This system and method provide efficient parallel training of a neural network model on multiple graphics processing units. The training module reduces the time and communication overhead of gradient accumulation and parameter updating of the network model in a neural network by overlapping processes in an advantageous way. In a described embodiment, the training module overlaps backpropagation, gradient transfer, and accumulation in a Synchronous Stochastic Gradient Descent algorithm on a convolutional neural network. The training module collects gradients of multiple layers during backpropagation of training from a plurality of graphics processing units (GPUs), accumulates the gradients on at least one processor, and then delivers the gradients of the layers to the plurality of GPUs during the backpropagation of the training. The whole model parameters can then be updated on the GPUs after receipt of the gradient of the last layer. Imai Haruki has 1 patent to his name.

Career Highlights

Imai Haruki is currently employed at International Business Machines Corporation, commonly known as IBM. His work at IBM focuses on advancing technologies related to artificial intelligence and machine learning. His contributions have been instrumental in enhancing the efficiency of neural network training processes.

Collaborations

Imai has collaborated with notable colleagues, including Tung Duc Le and Yasushi Negishi. These collaborations have fostered innovation and have led to advancements in their respective fields.

Conclusion

Imai Haruki is a key figure in the realm of neural network training, with a focus on improving efficiency through innovative methods. His work at IBM and his patent contributions highlight his commitment to advancing technology in artificial intelligence.