Responsible AI: Fairness and Bias Mitigation in Machine Learning

Instructor: Mia

Course Overview

• Duration: 3 days
• Structure: 4 modules per day

Learning Outcomes

• Theoretical Understanding
  • Basics of machine learning
  • Origins of bias in machine learning
• Hands-On Application
  • Practical machine learning skills
  • Training, tuning, testing, and evaluating models
  • Checking for and mitigating bias

Day 1: Modules

1. Fundamentals of Machine Learning
2. Introduction to Fairness and Bias Mitigation in ML
3. Model Formulation and Data Collection
4. Exploratory Data Analysis

• Objective: Cover the ML lifecycle from ideation to production while identifying and mitigating biases.

Day 2: Modules

1. Data Processing
2. Machine Learning Algorithm Selection
3. Model Building and Evaluation
4. Deeper Dive into Fairness Criteria

• Focus: Bias mitigation during pre-processing.

Day 3: Modules

1. Bias Mitigation during Model Training
2. Bias Mitigation during Post-Processing
3. Handling Bias in Production Models
4. Explainability

• Objective: Explaining model results to stakeholders, customers, and users.