Deliver "Smarter" Faster: Design Methodology for AI/ML Processor Design
By Joe Sawicki, executive vice president of IC EDA division, Mentor
EETimes (June 19, 2019)
New design tools can develop circuits for AI and machine learning faster than ever. AI/ML is being used to make those same design tools even faster.
We are at the beginning of an age where artificial intelligence (AI) processing will advance in sophistication rapidly and become ubiquitous. While the concept of AI — giving machines the ability to mimic cognitive functions to learn and solve problems and then take an action — has been an academic discipline since the mid-1950s, it wasn’t until the last five years that AI processing, mostly in the form of machine learning (ML), could step out of the dimly-lit halls of research and supercomputer one-offs and move to practical everyday use. Why?
The amount of data generated from the internet and billions of smart devices alone has given us more than enough data to collect sizable data sets with which we can have ML filter and train ML-based systems to use. In addition, today we have enough ubiquitous high-performance compute power in smart devices and the high-bandwidth communications infrastructure to process and transfer massive data sets quickly. This compute power also gives us the canvas to develop ever more sophisticated and specialized algorithms for particular tasks, further expanding the application of AI/ML.
To read the full article, click here
Related Semiconductor IP
- Zigbee Transceiver PHY
- Data Flow Architecture IP
- AMBA SPI Controller MRAM Controller
- Ethernet MAC
- Protocol Bridges
Related Articles
- Physical Design Exploration of a Wire-Friendly Domain-Specific Processor for Angstrom-Era Nodes
- VLSI Physical Design Methodology for ASIC Development with a Flavor of IP Hardening
- The role of cache in AI processor design
- Soft Tiling RISC-V Processor Clusters Speed Design and Reduce Risk
Latest Articles
- A Low-Latency ASIC Architecture for Real-Time Line Segment Detection
- BitFair: A 12nm Bit-Serial CNN Accelerator with Learnable Early Termination and Adaptive Bit Ordering for Ultra-Low-Power XR Vision
- A Flexible Sparsity-Aware FPGA Accelerator with Column-Wise Compression for Efficient CNN Inference
- Reducing Instruction-Fetch Energy in RISC-V for Embedded AI Processing via Dynamic and Static Loop Caching
- SPARC: Automated Root-Cause Analysis of Pre-Silicon Power Side-Channel Leakage in the Processor Design Flow