P041Filed

Ternary Tensor Decomposition and Factorization

The compiler knows the hardware — scheduling ternary execution across heterogeneous cores.

AU Application
2023900041
Filing Date
5 August 2023
Index Number
P041
Figures
10 figures
Batch / Category
Core 2

Explore the Vision

Discover this technology through five complementary perspectives — from technical architecture to partnership outcomes. Each layer reveals a different aspect of how this innovation creates value.

The compiler knows the hardware — scheduling ternary execution across heterogeneous cores.

What It IS

Technical Vision

The architectural essence — what makes this technology work

A compiler's view of a heterogeneous chip — ternary cores, conventional cores, memory banks, buses — all coordinated by an intelligent scheduler that maps ternary operations to their optimal hardware home. The software understands the silicon.

1/5
Explore the buyer's journey across 5 perspectives

Abstract

Low-rank tensor decomposition methods for further compressing ternary models while maintaining inference accuracy within specification.

Visual Essence

A compiler's view of a heterogeneous chip — ternary cores, conventional cores, memory banks, buses — all coordinated by an intelligent scheduler that maps ternary operations to their optimal hardware home. The software understands the silicon.

Visual Family:silicon-blueprint

Technology Domains

Related Patents

From the silicon-blueprint visual family