Training on Machine Learning Fundamentals for Engineers

Module 1: Engineering Thinking Frameworks for Machine Learning

• Applying engineering principles to machine learning problems
• Understanding the system-level view of ML systems
• Developing engineering judgement for ML design choices
• Building robust, reliable, and maintainable ML systems
• Integrating ML with existing engineering workflows
• Case Study: Applying engineering thinking to an ML problem

Module 2: Generative Artificial Intelligence and Large Language Models

• Understanding generative AI and its engineering implications
• Exploring large language models and their architecture
• Engineering considerations for LLM deployment and scaling
• Addressing performance, cost, and reliability in LLMs
• Building engineering solutions with generative AI
• Case Study: Engineering a solution with generative AI

Module 3: Generative Modelling: Physics-Based, Hybrid, and Data-Driven Approaches

• Understanding physics-based modelling and its applications
• Exploring hybrid approaches combining physics and data
• Engineering data-driven generative models
• Integrating generative models with physical systems
• Evaluating and validating generative models
• Case Study: Building a hybrid generative model

Module 4: Agentic Artificial Intelligence Concepts and Applications

• Understanding agentic AI and its engineering implications
• Exploring agent architectures and design patterns
• Engineering multi-agent systems for complex tasks
• Addressing coordination, communication, and learning in agents
• Building engineering solutions with agentic AI
• Case Study: Engineering an agentic AI solution

Module 5: Supervised and Semi-Supervised Learning

• Understanding supervised learning and its applications
• Exploring key supervised learning algorithms and techniques
• Applying semi-supervised learning for data efficiency
• Engineering supervised learning systems
• Addressing challenges in supervised learning: overfitting, bias
• Case Study: Building a supervised learning system

Module 6: Unsupervised Learning and Representation

• Understanding unsupervised learning and its applications
• Exploring clustering, dimensionality reduction, and representation learning
• Engineering unsupervised learning systems
• Addressing challenges in unsupervised learning: interpretability
• Integrating unsupervised learning into engineering workflows
• Case Study: Building an unsupervised learning system

Module 7: Reinforcement Learning Fundamentals

• Understanding reinforcement learning concepts and frameworks
• Exploring key reinforcement learning algorithms and techniques
• Engineering reinforcement learning systems
• Addressing challenges in reinforcement learning: exploration, credit assignment
• Building engineering applications with reinforcement learning
• Case Study: Building a reinforcement learning system

Module 8: Integration of Physics-Based and Data-Driven Models

• Understanding the benefits and challenges of integration
• Developing hybrid modelling approaches
• Engineering integrated models for real-world applications
• Validating and verifying integrated models
• Addressing uncertainty and error propagation
• Case Study: Building an integrated physics-data model

Module 9: Causality, Uncertainty, Validation, and Reproducibility

• Understanding causality in machine learning systems
• Quantifying and managing uncertainty in ML predictions
• Validating ML models for reliability and trustworthiness
• Ensuring reproducibility in ML research and development
• Building robust and trustworthy ML systems
• Case Study: Validating and ensuring reproducibility in ML

Module 10: Responsible Machine Learning Practice and Ethical Considerations

• Understanding the ethical dimensions of machine learning
• Addressing bias, fairness, and accountability in ML
• Developing responsible ML practices and guidelines
• Building ethical and trustworthy ML systems
• Engaging with stakeholders on responsible ML
• Case Study: Building a responsible ML system