Overview

This course will give you an overview on the concept of AI privacy, which helps in building trust in AI, and explains how open source tools from IBM can help both assess the privacy risk of AI-based solutions, and help them adhere to any relevant privacy requirements.  Learners will start off with an overview of the AI Privacy concept, and then do a deep dive into three IBM open source tool kits: AI Privacy Toolkit, Differential Privacy Library and Adversarial Robustness Toolbox which help assess and create machine learning models that preserve the privacy of their training data and comply with relevant data protection regulations.

Audience

Analytics Leaders, Data Science Leaders, Practicing Data Scientists, Machine Learning Engineers, AI specialists. Anyone with an interest in AI Trust and Privacy having the prerequisite knowledge required.

Prerequisites

Students should have a basic understanding of:

• AI/Machine Learning Workflow
• Data Science
• Python

Objective

In this course you will learn:

• Setting up and setting the scene with IBM Watson Studio
• Gathering and accessing data for the team with IBM Watson Studio
• Prototyping with IBM Watson Studio Collaborative Notebooks

Course Outline

Setting up and setting the scene with IBM Watson Studio

• Describe the use case
• List the steps in the Data Science workflow
• Add Watson Studio as a service from an existing IBM Cloud account

Gathering and accessing data for the team with IBM Watson Studio

• Create a project in Watson Studio
• Invite collaborators and assign permissions
• Add data to the project from a local file

Prototyping with IBM Watson Studio Collaborative Notebooks

• Open a notebook with their desired environment
• Generate code to import data in the notebook
• Save notebook versions
• Share a notebook
• Describe the steps in the Data Science window