Company Filing History:
Years Active: 2022
Title: Innovations by Prasanth Chatarasi
Introduction
Prasanth Chatarasi is an accomplished inventor based in Atlanta, GA. He has made significant contributions to the field of tensor processing, particularly through his innovative patent. His work focuses on enhancing the efficiency of tensor convolutions, which are crucial in various computational applications.
Latest Patents
Chatarasi holds a patent titled "Generation of vector codes for tensor convolutions." This patent discloses approaches for generating vector codes that involve inputting tensor processing statements. Each statement specifies an output variable, an initial variable, and multiply-and-accumulate (MAC) operations. The MAC operations are organized into groups, with each group referencing the same output variable and having overlapping references to elements of the first tensor. For each group, instructions are generated to load elements of the first tensor into a first register and elements of the second tensor into a second register. This innovative approach streamlines the process of selecting elements for input to an array of MAC circuits.
Career Highlights
Chatarasi is currently employed at Xilinx, Inc., where he continues to develop cutting-edge technologies in the field of tensor processing. His work has positioned him as a key player in the advancement of computational efficiency.
Collaborations
Throughout his career, Chatarasi has collaborated with notable colleagues, including Stephen Andrew Neuendorffer and Samuel R Bayliss. These collaborations have further enriched his contributions to the field.
Conclusion
Prasanth Chatarasi's innovative work in tensor processing exemplifies the impact of modern inventions on computational technology. His patent on generating vector codes for tensor convolutions showcases his expertise and commitment to advancing the field.