Introduction to Neural Networks
Begin with the fundamentals. This course covers perceptrons, multilayer networks, backpropagation, and gradient descent through hands-on implementation.
- Build networks from scratch in Python
- Understand forward and backward passes
- Implement common activation functions
- Debug training convergence issues
- Work with real classification datasets
Advanced Deep Learning
Master modern architectures and techniques used in production systems. Covers transformers, attention mechanisms, normalization strategies, and advanced optimization.
- Implement transformer architectures
- Understand self-attention mechanisms
- Apply layer normalization and batch normalization
- Use advanced optimizers like Adam and AdamW
- Fine-tune pretrained models
Computer Vision with CNNs
Learn to build systems that interpret visual information. From basic convolutional layers to state-of-the-art architectures for object detection and segmentation.
- Understand convolutional and pooling operations
- Implement ResNet and other modern architectures
- Train models for image classification
- Apply transfer learning techniques
- Build object detection pipelines
Natural Language Processing
Process and generate text using neural approaches. Covers embeddings, sequence models, transformers, and practical applications like sentiment analysis and text generation.
- Create and use word embeddings
- Build recurrent and transformer-based models
- Implement attention for sequence-to-sequence tasks
- Fine-tune language models for specific domains
- Handle tokenization and preprocessing
Reinforcement Learning
Train agents that learn through interaction with environments. Explore value-based and policy-based methods through game simulations and control tasks.
- Implement Q-learning and deep Q-networks
- Understand policy gradient methods
- Build actor-critic architectures
- Apply reinforcement learning to games
- Handle exploration vs exploitation tradeoffs
AI Project Capstone
Apply everything you've learned to a substantial project. Work with mentorship to design, implement, and present a complete AI system of your choosing.
- Define project scope and objectives
- Design appropriate architectures for your problem
- Implement and train models
- Evaluate and iterate on performance
- Present findings and technical decisions
