Patent US 11,080,813

Patent №

US 11,080,813

Granted

—

Owner

—

Lab

—

AI components

hardware

Assignment

None on record

Dataset

AIPD

2023_r1 edition

Application

16584076

View graph

An apparatus to facilitate compute optimization is disclosed. The apparatus includes a mixed precision core to perform a mixed precision multi-dimensional matrix multiply and accumulate operation on 8-bit and/or 32 bit signed or unsigned integer elements.

AI hardwareG06T 1/20G06F 3/14G06F 9/3001G06F 9/30014G06F 9/3017G06F 9/3851G06F 9/3887G06F 9/3888+16 more

Google Patents

AI classification

AI hardware1.00

Vision0.02

Knowledge representation0.00

Natural language0.00

Speech0.00

Evolutionary computation0.00

Machine learning0.00

Planning0.00

Cites (7)

SPARSE CONVOLUTIONAL NEURAL NETWORK ACCELERATORUS 10,891,538 · 2021 SPARSE CONVOLUTIONAL NEURAL NETWORK ACCELERATORUS 10,860,922 · 2020 SPARSE CONVOLUTIONAL NEURAL NETWORK ACCELERATORUS 10,528,864 · 2020 FUSED MULTIPLY-ADD FUNCTIONAL UNITUS 8,106,914 · 2012 FUSED MULTIPLY ADD SPLIT FOR MULTIPLE PRECISION ARITHMETICUS 7,720,900 · 2010 DATA PROCESSING SYSTEM AND METHOD FOR PERFORMING AN ARITHMETIC OPERATION ON A PLURALITY OF SIGNED DATA VALUESUS 6,338,135 · 2002 Patent US 7,873,812US 7,873,812

Related research

MIX-ACIM: A 28-nm Mixed-Precision Analog Compute-in-Memory With Digital Feature Restoration for Vector-Matrix MultiplicationOpenAlex A Unified Programmable Edge Matrix Processor for Deep Neural Networks and Matrix AlgebraSemantic Scholar The Cambrian Explosion of Mixed-Precision Matrix Multiplication for Quantized Deep Learning InferencearXiv Precise and scalable analogue matrix equation solving using resistive random-access memory chipsOpenAlex Hardware-efficient matrix multiplication core optimization for edge AI on FPGAOpenAlex

Full text

Google Patents