With the help of advanced algorithms and computing hardware (GPU), deep learning has pushed artificial intelligence to a new level. Thousands of parallel processing ALUs make GPUs powerful machines that can perform matrix multiplication for DNN operations. By sacrificing flexibility, DNN acceleration chips built with ASICs like TPUs can achieve higher performance and lower power consumption. However, using digital circuits for matrix multiplication has its limitations. To achieve higher acceleration factors and lower power consumption, the in-store computing IMC approach for vector matrix multiplication for DNNs has been proposed. While IMC brings great benefits for DNNs, it also faces many challenges. For example, deficiencies in the storage unit, data transfer between layers, and supporting software and frameworks need to be overcome before IMC DNNs can become a reality.
Founded in August 2020, WIMI Hologram Academy is dedicated to holographic AI vision exploration, and conducts research on basic science and innovative technologies, driven by human vision. The Holographic Science Innovation Center, in partnership with WIMI Hologram Academy, is committed to exploring the unknown technology of holographic AI vision, attracting, gathering and integrating relevant global resources and superior forces, promoting comprehensive innovation with scientific and technological innovation as the core, and carrying out basic science and innovative technology research.
Contacts
Holographic Science Innovation Center
Email: pr@holo-science. com
