Does Your AI Chip Have Its Own DNN?
By Junko Yoshida, EETimes
August 25, 2019
For AI accelerators in the race to achieve optimum accuracy at minimum latency, especially in autonomous vehicles (AVs), teraflops have become the key element in many so-called brain chips. The contenders include Nvidia’s Xavier SoC, Mobileye’s EyeQ5, Tesla’s Full Self-Driving computer chip and NXP-Kalray chips.
In an exclusive interview with EE Times last week, Forrest Iandola, CEO of DeepScale, explained why this sort of brute-force processing approach is unsustainable, and said many of the assumptions common among AI hardware designers are outdated. As AI vendors gain more experience with more AI applications, it's becoming evident to him that different AI tasks are starting to require different technological approaches. If that's true, the way that AI users buy AI technology is going to change, and vendors are going to have to respond.
Rapid advancements in neural architecture search (NAS), for example, can make the search for optimized deep neural networks (DNN) faster and much cheaper, Iandola argued. Instead of relying on bigger chips to process all AI tasks, he believes there is a way “to produce the lowest-latency, highest-accuracy DNN on a target task and a target computing platform.”
To read the full article, click here
Related Semiconductor IP
- NFC wireless interface supporting ISO14443 A and B with EEPROM on SMIC 180nm
- DDR5 MRDIMM PHY and Controller
- RVA23, Multi-cluster, Hypervisor and Android
- CXL 3.0 Controller
- ECC7 Elliptic Curve Processor for Prime NIST Curves
Related News
- Skymizer Launches HyperThought: Build Your Own AI Chip with Skymizer’s LPU IP
- Does Huawei have its eye on Nokia?
- Does NVMe Have a Place in Industrial Embedded and IoT?
- Does China Have Imagination?
Latest News
- AIStorm and DB HiTek Debut SpectroMic™ KWS—an 18uA Always-on Keyword-Spotting Solution Enabling IoT AI Voice Interaction
- SignatureIP Unveils Industry-Leading CXL 3.2 Solution for High-Performance Computing
- Synopsys and Ansys Provide Update Regarding Expected Timing of Acquisition Close
- CAST Releases First Dual LZ4 and Snappy Lossless Data Compression IP Core
- Arteris Wins “AI Engineering Innovation Award” at the 2025 AI Breakthrough Awards