Darsen Lu《New Era for AI Chips》
Summary With the rapid advancement of artificial intelligence and IoT, many solutions have been successfully implemented. However, for very large biomedical image computations, such as MRI’s, the deep learning/ training will be a lot more time consuming. At the moment, GPU’s are considered the baby of the AI world, Google has also developed a TPU to expedite AI procedures. Both GPU and TPU make the process swift. Prof. Darsen Lu’s Research Team has built an open simulation platform called (simNemo) to enable design exploration by academic and industry users in Taiwan on both the AI platform and AI application fronts.
It essentially overcomes the von-Neumann bottleneck wherein computation is limited by the transfer of data from memory to the CPU. With the use of neuromorphic technology, the goal is for less energy consumption as well as expedited deep learning. This platform will eventually be able to assist both industry and academia in the manufacturing of software or app chips to meet their needs.
Keyword AI中心 成功大學 類神經電路驗證 AI晶片 半導體開發
Provide the latest information of AI research centers and applied industries