P012Filed

Ternary Inference for Medical Image Analysis

Privacy-preserving collaborative training across the planet — federated ternary learning.

AU Application
2023900012
Filing Date
10 March 2023
Index Number
P012
Figures
9 figures
Batch / Category
Applications

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.

Privacy-preserving collaborative training across the planet — federated ternary learning.

What It IS

Technical Vision

The architectural essence — what makes this technology work

Edge devices across the globe — phones, sensors, vehicles — each training a local ternary model on private data, sharing only ternary gradient updates with the collective. No raw data leaves any device. A planet learning together while keeping secrets.

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

Abstract

Deployment of ternary neural networks for radiological image segmentation and classification, achieving 98% accuracy with 50% latency reduction on edge devices.

Visual Essence

Edge devices across the globe — phones, sensors, vehicles — each training a local ternary model on private data, sharing only ternary gradient updates with the collective. No raw data leaves any device. A planet learning together while keeping secrets.

Visual Family:edge-bloom

Technology Domains

Related Patents

From the edge-bloom visual family