This inventor holds 2 USPTO granted patents and 1 published patent application. Top assignee: International Business Machines Corporation. Active years: 2018-2019.
Company Filing History:
Years Active: 2018-2019
Title: Arash Ashari: Innovator in Machine Learning Optimization
Introduction
Arash Ashari is a prominent inventor based in Kirkland, WA (US). He has made significant contributions to the field of machine learning, particularly in optimizing workloads on graphical processing units (GPUs). With a total of 2 patents, Ashari's work is paving the way for advancements in computational efficiency.
Latest Patents
Arash Ashari's latest patents focus on a pipelined approach to fused kernels for the optimization of machine learning workloads on GPUs. The first patent describes a method that identifies computations with generic patterns commonly observed in machine learning processes. This method includes hierarchical aggregation spanning the memory hierarchy of the GPU, which maintains partial output vector results in shared memory. The second patent also emphasizes the optimization of machine learning workloads, employing an optimized fused GPU kernel to exploit temporal locality for inherent data-flow dependencies. This innovative approach enhances the efficiency of GPU kernel launch parameters, maximizing thread occupancy while minimizing atomic writes to GPU global memory.
Career Highlights
Ashari is currently employed at International Business Machines Corporation (IBM), where he continues to develop cutting-edge technologies in machine learning and GPU optimization. His work has garnered attention for its potential to significantly improve the performance of machine learning applications.
Collaborations
Throughout his career, Arash has collaborated with notable colleagues, including Matthias Boehm and Alexandre Valentinovich Evfimievski. These partnerships have contributed to the advancement of research and innovation in the field.
Conclusion
Arash Ashari's contributions to machine learning optimization through his patents and work at IBM highlight his role as a key innovator in the technology sector. His research is instrumental in enhancing the efficiency of machine learning processes on GPUs, showcasing the importance of innovation in this rapidly evolving field.