Jens Huthmann

Title: Innovations of Jens Huthmann

Introduction

Jens Huthmann is a notable inventor based in Hyogo, Japan. He has made significant contributions to the field of technology, particularly in the area of neural networks and hardware acceleration. With a total of three patents to his name, Huthmann continues to push the boundaries of innovation.

Latest Patents

Huthmann's latest patents include groundbreaking work in neural network hardware acceleration. One of his patents focuses on data parallelism, which is performed by an integrated circuit that includes multiple memory banks. Each memory bank is designed to store values and transmit them efficiently. Additionally, the circuit features several computation units, each equipped with either a channel pipeline or a multiply-and-accumulate (MAC) element. These units perform mathematical operations on input data values and weight values to generate resultant data values. Another significant patent involves sequence partition-based schedule optimization. This process generates a sequence and a schedule based on that sequence, dividing it into multiple partitions according to a data dependency graph. Each partition includes a portion of instructions and buffers, and the method involves optimizing these partitions before merging them into a cohesive schedule.

Career Highlights

Huthmann is currently employed at Edgecortix Pte. Ltd., where he applies his expertise in developing advanced technologies. His work has been instrumental in enhancing the efficiency and performance of neural network applications.

Collaborations

Some of Huthmann's notable coworkers include Nikolay Nez and Sakyasingha Dasgupta. Their collaborative efforts contribute to the innovative environment at Edgecortix Pte. Ltd.

Conclusion

Jens Huthmann is a prominent inventor whose work in neural network hardware acceleration and optimization continues to influence the technology landscape. His patents reflect a commitment to advancing the field and improving computational efficiency.