All Courses
Course Syllabus

Advanced Deep Learning & Systems

Frontier architectures & systems: Diffusion Models, Transfer Learning, PEFT/LoRA, PyTorch Internals, Deployment/ONNX, Multimodal AI, and Mixture of Experts.

9Notebooks
Start Lesson 1
Lesson #1
Advanced

23. Diffusion Models & Generative Denoising (DDPM)

Mathematical foundations and implementations of Denoising Diffusion Probabilistic Models (DDPM): forward Gaussian noise scheduling, reverse denoising U-Net, Score-based models, and latent diffusion.

Diffusion ModelsDDPM+4
22 min
Read Notebook
Lesson #2
Intermediate

24. Transfer Learning & Domain Adaptation

Leveraging pretrained representations: feature extraction vs full fine-tuning, strategic layer freezing, learning rate warmups, and domain adaptation techniques.

Transfer LearningPretrained Models+4
18 min
Read Notebook
Lesson #3
Advanced

25. Parameter-Efficient Fine-Tuning (PEFT, LoRA & QLoRA)

State-of-the-art parameter-efficient adaptation: Low-Rank Adaptation (LoRA) matrix decomposition, rank selection, alpha scaling, 4-bit NormalFloat QLoRA, and Prefix Tuning.

PEFTLoRA+5
22 min
Read Notebook
Lesson #4
Intermediate

26. Deep Learning Frameworks (PyTorch Core Architecture)

Mastering the PyTorch engine: tensor memory layouts, autograd computational graphs, custom nn.Module building blocks, dataset/DataLoader pipelining, and multi-GPU training.

PyTorchTensors+5
30 min
Read Notebook
Lesson #5
Intermediate

27. Model Training, Validation & Evaluation Pipelines

Production training methodologies: train/val/test data leakage prevention, cross-validation, learning rate schedulers, early stopping, and metric evaluation (ROC-AUC, F1, PR curves).

Training LoopsValidation+4
25 min
Read Notebook
Lesson #6
Advanced

28. Deep Learning Deployment, ONNX & Inference Optimization

Deploying models to production: serialization checkpoints, TorchScript JIT tracing, ONNX graph export, INT8/FP16 quantization, TensorRT acceleration, and low-latency serving.

DeploymentONNX+5
24 min
Read Notebook
Lesson #7
Advanced

29. Large Language Models (LLM) Pre-training & Inference

End-to-end foundation model engineering: autoregressive pre-training datasets, byte-pair tokenization, causal self-attention, KV caching optimization, and decoding strategies (top-p, temperature).

LLMsPre-training+5
26 min
Read Notebook
Lesson #8
Advanced

30. Multimodal Deep Learning & Vision-Language Models

Bridging vision and language: OpenAI CLIP contrastive pre-training, modality-specific encoders, cross-attention projection bottlenecks, and visual instruction tuning.

MultimodalVision-Language+4
24 min
Read Notebook
Lesson #9
Advanced

31. Advanced Paradigms (Mixture of Experts & Scaling Laws)

Frontier deep learning architectures: Mixture of Experts (MoE) sparse routing, top-k expert gating, load balancing auxiliary loss, Chinchilla scaling laws, and speculative decoding.

Mixture of ExpertsMoE+4
32 min
Read Notebook