Day 1: Foundations of AI and Machine Learning
Overview of AI and ML Concepts
- History, current trends, and applications across industries.
- Key terminology, methodologies, and problem-solving approaches.
- Introduction to Machine Learning Types:
- Supervised, unsupervised, and reinforcement learning.
- Comparison and use cases.
Key Terminology
- Features, labels, training data, and test data.
- Concepts of overfitting and underfitting.
Tools and Frameworks
- Overview of popular tools: Python, TensorFlow, PyTorch, and Google AutoML.
Day 2: Data Analysis and Preprocessing
Data Preparation Techniques
- Cleaning, transforming, and normalising datasets.
- Visualisation techniques for understanding data patterns.
- Tools and libraries for visualisations (e.g., Python, NumPy, Pandas).
Supervised Learning Algorithms
- Regression techniques (linear and polynomial regression).
- Gradient Descent: Revolutionary technique in modern AI.
- Classification techniques (Nearest Neighbours, SVM).
- Decision Trees and Random Forests.
Practical Session
- Use Python and IBM SPSS for predictive modelling.
- Work with structured datasets (e.g., sales data, employee performance, financial data).
- Analyse feature importance and optimise the model.
Day 3: Deep Learning and Unsupervised Learning
Unsupervised Learning Algorithms
- Clustering (K-means, Hierarchical).
- Content-based filtering.
- Practical applications of unsupervised learning techniques.
Deep Learning Fundamentals
- Neural networks, activation functions, and backpropagation.
- Architectures: Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN).
- Applications of transfer learning.
Practical Session
- Apply K-Means clustering to real-life datasets.
- Train and fine-tune a deep learning model using Google AutoML Vision for object detection.
- Explore unsupervised learning using Google AutoML Tables.
Day 4: Natural Language Processing (NLP) and Automation
NLP Fundamentals
- Tokenization, embeddings, and language models.
- Applications in sentiment analysis and Named Entity Recognition (NER).
- Large Language Models (LLMs).
Practical Session and Tasks
Custom GPT Configuration:
- LLM development using GPT.
- Overview of OpenAI GPT fine-tuning.
- Steps to create a custom GPT for specialized business use cases.
- Example use case: Chatbot for customer inquiries.
Workflow Automation:
- Using Zapier with OpenAI for automated processes.
- Building workflows (e.g., email summaries, task automation).
Day 5: Deployment, Ethical Implications, and Future of AI
Deploying AI Models
- Basics of deploying ML models in cloud platforms.
- Introduction to explainability and interpretability in AI.
- Interpretability and explainability of ML models.
Ethical Implications
- Fairness, bias, and ethical considerations in AI.
- Privacy and security concerns.
- Regulatory frameworks and compliance.
Capstone Project
- Design a solution using learned techniques and develop an AI roadmap for your company (e.g., predictive modelling or NLP workflows).
