In this Intel Chip Chat audio podcast with Allyson Klein: Dr. Pradeep Dubey, Intel Fellow at Intel Labs, joins us live from Intel Developer Forum in San Francisco. Dr. Dubey’s Parallel Computing Lab focuses on compute intensive applications such as machine learning. Dr. Dubey discusses the need to scale infrastructure to meet the growing demands of AI. Where software was once designed to make the right decisions, today it increasingly is designed to ask the right questions. Answering those questions using machine learning can require significant computational resources. Dr. Dubey highlights the ways Intel is accelerating machine learning via projects like the Intel Math Kernel Library for Deep Neural Networks (Intel MKL-DNN), an open source library for deep learning, and Intel-optimized Caffe, a fork dedicated to improving the framework’s performance when running on CPUs, especially Intel Xeon processors.

For more information on Dr. Dubey’s work, please visit:
pcl.intel-research.net

Tags: Allyson Klein, Audio Podcast, Chip Chat, Deep Neural Networks, Dr. Pradeep Dubey, Intel, Intel Labs, Intel Math Kernel Library, Intel MKL-DNN, Intel Xeon Processors, machine learning
 
Posted in: Audio Podcast, Intel, Intel Chip Chat