While 2015 and 2016 were, in many ways, breakthrough years for Artificial Intelligence (AI) and Machine/Deep Learning in particular, most of the focus has been on developing better algorithms. AI, however, can also benefit greatly from better hardware which is becoming a big differentiator in the performance of AI networks.
GPUs and AI-specific processors have the potential to displace the traditional CPU in the next few years with profound implications for industry stalwarts such as Intel. This report compares and contrasts the various alternative processor architectures for both the server market and client-side devices, and includes short profiles of leading processor vendors and their technologies.
Table of Contents:
- Strategic Planning Assumptions
- Machine Learning
- Processor Options for Machine Learning
- AI Chip Applications
- Key Players
- General Purpose OR Specialized AI chips?
This report can be read in conjunction with the presentation:
- AI at the Edge: Deep Learning Sparks Innovation for Chip Vendors