Company Filing History:
Years Active: 2024-2025
Title: **James Michael Bodwin: Innovator in Matrix Multiplication and Neural Networks**
Introduction
James Michael Bodwin, an accomplished inventor based in Cupertino, California, has made significant contributions to the fields of mathematics and artificial intelligence. With a total of four patents to his name, Bodwin's innovations focus on improving computational efficiency and reducing power consumption in complex algorithms.
Latest Patents
Among his notable patents, two stand out for their innovative approach to technology:
1. **Methods for multiplying matrices using a plurality of chiplets**: This patent introduces a method where a plurality of chiplets can efficiently multiply two matrices, A and B. In this technique, matrix A is decomposed into horizontal stripes, while matrix B is decomposed into vertical stripes. By allowing each horizontal stripe to multiply with every vertical stripe, the output matrix C is formed. The design optimally stores horizontal stripes in a distributed manner across the chiplets, while vertical stripes are passed between chiplets ensuring each receives the necessary data for processing.
2. **Low power hardware architecture for a convolutional neural network (CNN)**: This patent presents a dynamic data quantization method aimed at minimizing power consumption in CNN systems. The innovation focuses on using quantized representations of activation values and kernels, streamlining the computation process in CNNs by significantly reducing the required power without sacrificing performance.
Career Highlights
Throughout his career, James Bodwin has been associated with prominent technology companies, including Recogni Inc. and Persimmons, Inc. His work has consistently revolved around leveraging advanced computational techniques to enhance machine learning applications and optimize hardware architecture.
Collaborations
Bodwin has collaborated with fellow innovators such as Jian Hui Huang and Pradeep R Joginipally, working together to further advancements in technology and exploring new frontiers in computational efficiency and neural network design.
Conclusion
James Michael Bodwin exemplifies the spirit of innovation in technology, focusing on matrix computations and neural network efficiency. His patents not only reflect his technical expertise but also contribute significantly to advancing machine learning methodologies, paving the way for future innovations in the tech industry.