eIQ® ML Watermarking Model Protection Tool Demo | NXP Semiconductors

デザイン・センター
トレーニング
eIQ® ML Watermarking Model Protection Tool Demo

eIQ® ML Watermarking Model Protection Tool Demo

アカウントが必要です

サインインこのコンテンツおよびサイトの追加機能にアクセスすることができます。アカウントをお持ちでないですか？今すぐご登録下さい。

About This Training

This training provides a demonstration of NXP eIQ® Watermarking Model protection tool within the eIQ Toolkit that helps developers prove ownership and detect copying of their model intellectual property.

Watermarking is the process of embedding information in the content, and the embedded information may not be apparent upon normal observation. With the eIQ® Model Watermarking tool addition into the NXP eIQ Toolkit for machine learning development, watermarking has also found its way into ML.

The watermark is based on a creative element — a secret drawing — which adds a piece of copyright-protected information to the ML model and provides an additional benefit to the NXP eIQ Model Watermarking tool. This secret drawing adds a piece of copyright-protected information to the ML model. This secret drawing helps strengthen a copyright claim toward any potential copyist.

Presentation

eIQ® ML Watermarking Model Protection Tool Demo

What You Will Learn

Demonstrates the eIQ Watermarking Model protection hidden functionality to add machine learning model images with a secret overlay
How to use the tool to help prove ownership of a machine learning model
How to use the tool to add copyright protection to a machine learning model
Demonstrates the addition of these functions without a loss of accuracy or performance

Presented by

Geert Derks
CTO Security Concepts Team, NXP Semiconductors

期間

7 min

言語

英語

スキルレベル

中級

0閲覧数4いいね

We are looking for ways to improve.

アンケートに答える。

Additional Resources

Application Software Pack: ML-Based System State Monitor

Related Trainings

NXP AI and Machine Learning Training Academy

Stay In the Loop

Subscribe to our newsletter to stay updated with our latest developments and if you need further assistance, we are here to help.

Subscribe Contact Support