Join Amit Kulkarni, an industry expert and certified AI specialist, as he walks you through the Mastering GenAI course at Sunbeam Pune. This program is designed to help you understand, build, and innovate with AI-powered solutions.

Week 1: Foundations of Machine Learning

Day 1: Introduction to Machine Learning

• What is Artificial Intelligence (AI) and Machine Learning (ML)?
  • AI vs. ML vs. Deep Learning
  • Real-world applications of ML
• Types of ML: Supervised, Unsupervised, and Reinforcement Learning.
• Overview of the ML Workflow: Data, Model, Training, and Evaluation.

 Day 2: Understanding Data and Features

• What is data?
  • Structured vs. Unstructured Data.
• Data Preprocessing: Cleaning and Transforming Data.
• Introduction to Features and Feature Engineering.
• Hands-on: Explore and preprocess a simple dataset using Python (e.g., Iris Dataset).

Day 3: Key Machine Learning Algorithms

• Linear Regression: Predicting numerical outcomes.
•  Decision Trees: Intuitive model for classification and regression.
• K-Means Clustering: Basics of unsupervised learning.
• Hands-on: Build and train a simple regression and classification model using SciKit-Learn.

Day 4: Model Evaluation

• Metrics for evaluating ML models:
  • Accuracy, Precision, Recall, F1 Score (classification).
  • Mean Squared Error (regression).
• Overfitting and Underfitting: Understanding Bias-Variance Tradeoff.
• Hands-on: Evaluate a simple ML model and tune hyperparameters.

Day 5: Introduction to Neural Networks

• Basics of Neural Networks:
  • Perceptrons, Layers, and Activation Functions.
  • Forward Pass and Backpropagation.
• Hands-on: Build and train a simple neural network.

Week 2: Deep Learning Essentials

Day 6: Deep Learning Fundamentals

• What is Deep Learning?
  • Differences between traditional ML and DL.
• Overview of Deep Neural Networks (DNNs).
• Key architectures: Feed-forward Neural Networks (FNNs).
• Hands-on: Build a multi-layer perceptron (MLP) for image classification (MNIST dataset).

Day 7: Convolutional Neural Networks (CNNs)

• Basics of CNNs:
  • Convolutions, Filters, and Pooling.
  • How CNNs excel in image processing tasks.
• Hands-on: Implement a CNN for image classification.

Day 8: Recurrent Neural Networks (RNNs)

• Basics of RNNs:
  • Understanding sequential data.
  • Introduction to Long Short-Term Memory (LSTM) networks.
• Applications: Time series forecasting, text processing.
• Hands-on: Build an RNN for text classification.

Day 9: Optimization and Training Techniques

• Gradient Descent Variants: SGD, Adam, etc.
• Regularization Techniques: Dropout, Batch Normalization.
• Hands-on: Improve CNN or RNN performance using regularization techniques.

Day 10: Introduction to Generative Models

• What is Generative AI?
  • Discriminative vs. Generative Models.
• Overview of Generative Models:
  • Variational Autoencoders (VAEs).
  • Generative Adversarial Networks (GANs).
• Hands-on: Visualize outputs of pre-trained generative models.

Week 3: Generative AI Deep Dive

Day 11: Variational Autoencoders (VAEs)

• Understanding VAEs:
  • Encoder-Decoder Architecture.
  • Latent Space and Sampling.
• Applications: Image generation, anomaly detection.
• Hands-on: Build a VAE for image reconstruction.

Day 12: Generative Adversarial Networks (GANs)

• How GANs work:
  • Generator and Discriminator.
  • Training dynamics and challenges.
• Applications: Image synthesis, style transfer.
• Hands-on: Implement a simple GAN to generate synthetic images.

Day 13: Transformers and Attention Mechanisms

• What are Transformers?
  • Attention Mechanism: Self-Attention and Multi-Head Attention.
  • Overview of GPT and BERT models.
• Applications: Text generation, language modelling.
• Hands-on: Use Hugging Face Transformers for text generation.

Day 14: Text-to-Text Applications

• Fine-tuning Pretrained Models:
  • Text summarization.
  • Question answering.
• Hands-on: Build a text summarizer using Hugging Face.

Day 15: Neural Style Transfer

• Understanding Style Transfer:
  • Content vs. Style Representation.
• Applications in Art and Design.
• Hands-on: Create AI-generated artwork using Neural Style Transfer.

Week 4: Advanced Topics and Applications

Day 16: Music and Video Generation

• Generating Music:
  • RNNs and Pre-trained Models (e.g., Magenta).
• Video Synthesis:
  • GANs for video generation.
  • Ethical concerns around deepfakes.
• Hands-on: Generate music using Magenta.

Day 17: Bias and Ethics in Generative AI

• Ethical Concerns:
  • Bias in AI models.
  • Misinformation through deepfakes.
• Responsible AI Practices.
• Class Discussion: Analyze examples of ethical issues in GenAI.

Day 18: Deployment of Generative AI Models

• Deployment Strategies:
  • Hosting models using Streamlit or Flask.
  • Using cloud platforms like AWS or Google Colab.
• Hands-on: Deploy a generative model with a simple web interface.

Day 19: Capstone Project Planning

• Choose a Project:
  • Text generation, image synthesis, or any real-world GenAI application.
• Plan:
  • Dataset preparation, tools, and expected outcomes.
• Team Collaboration: Work in groups to brainstorm project ideas.

Day 20: Capstone Project Development and Presentation

• Complete and Showcase the Project:
  • Build a working Generative AI application.
• Presentations:
  • Share the project, including challenges and learnings.
  • Peer reviews and feedback.