By the end of this module, learners will Understand how deep learning works at a fundamental level Be able to build and train basic neural networks from scratch Gain experience using major deep learning libraries Be prepared to tackle more advanced architectures like CNNs and RNNs

• Introduction to Deep Learning and AI vs ML vs DL
• Basics of Neural Networks (Perceptron, MLP)
• Activation Functions: ReLU, Sigmoid, Tanh
• Forward & Backward Propagation
• Loss Functions & Optimization (Gradient Descent, Adam)
• Introduction to Keras/TensorFlow or PyTorch

By the end of this module, learners will Understand how to design, train, and evaluate CNNs for visual data tasks Be able to implement and apply RNNs and their variants to sequential data problems Gain practical experience in using CNNs for image recognition and RNNs for tasks like language modeling or time-series prediction

• Convolution, Pooling, Filters, Padding
• CNN architectures: LeNet, AlexNet, VGG
• Basic RNNs, Vanishing Gradient Problem
• LSTM, GRU, Bidirectional RNNs
• Time Series Forecasting

By the end of this module, learners will Understand and apply regularization techniques to enhance deep learning models Use advanced model architectures and optimize them with state-of-the-art methods Apply transfer learning to achieve high performance on custom datasets with limited data Gain practical skills in deploying real-time AI models using pre-trained networks

• Dropout, Batch Normalization, Regularization
• Pre-trained Models: VGG, ResNet, Inception
• Fine-tuning and Feature Extraction
• Attention Mechanism Basics
• Deployment with Streamlit or Flask

By the end of this module, learners will: Understand the fundamentals of generative modeling Build and train autoencoders and GANs Generate new data samples such as realistic images Evaluate the quality of generative outputs using established metrics

• What is Generative AI?
• Autoencoders
• Vanilla, Sparse, Denoising, Variational Autoencoders (VAEs)
• GANs (Generative Adversarial Networks)
• Generator vs Discriminator
• DCGANs, Conditional GANs (cGAN)
• Evaluation Metrics: FID, Inception Score

By the end of this module, learners will Understand and implement the Transformer architecture Work with large-scale pre-trained language models for various NLP tasks Create and fine-tune models using Hugging Face Apply prompt engineering and API tools for powerful generative applications Gain initial exposure to state-of-the-art diffusion models in image generation

• Transformer Architecture: Self-Attention, Multi-head Attention
• Pre-trained Language Models
• BERT, GPT-2/3/4 basics
• Hugging Face Transformers
• Prompt Engineering 101
• Intro to Diffusion Models
• Text Summarization using BERT
• Chatbot using GPT-2 or GPT-3 via Hugging Face API
• Image generation using DALL·E (API-based)

By completing a capstone project, learners will Demonstrate their ability to build full-stack AI solutions Showcase mastery of model training, fine-tuning, and deployment Build a strong portfolio piece with real-world relevance Gain confidence presenting technical work to peers or employers

• AI Art Generator using Diffusion or GANs
• Personal Assistant Chatbot with GPT
• Text-to-Image Generation using Prompt + API
• Multimodal AI: Combine text + image inputs/outputs

By the end of the program, learners will Gain proficiency in using modern deep learning libraries Be comfortable deploying AI models using Flask or Streamlit Understand containerization and version control workflows Be well-prepared to work in real-world AI development environments

• Core Libraries: TensorFlow, PyTorch, Keras
• GenAI Tools: Hugging Face, OpenAI API, Stability AI, Replicate
• Visualization: Matplotlib, Seaborn, TensorBoard
• Deployment: Flask, Streamlit, Docker
• Others: Git, Colab, Jupyter, Weights & Biases