Authoritative Readings and Resources

These resources extend the supplied learning chapters in AINS6003 Deep Learning & Neural Networks. They were selected because they are primary standards, official documentation, open textbooks, or authoritative institutional guidance—not unsourced link lists.

How to Read Them

For each module, read the supplied chapter first. Then use the two linked resources at the end of that chapter to test terminology, compare the course’s worked example with an authoritative treatment, and identify one point that should change or qualify your recommendation. Students are not expected to read every linked document cover to cover.

1. PyTorch Tutorials

Official tutorials for tensors, autograd, models, training, and deployment.

Use with: From neurons to multilayer networks, Convolutional neural networks for vision, Sequence models: RNNs and LSTMs, GPU workflows, scale, and deployment.

2. Dive into Deep Learning

Open, executable deep-learning textbook.

Use with: From neurons to multilayer networks, Backpropagation and automatic differentiation, Sequence models: RNNs and LSTMs, Attention and transformers.

3. Stanford CS231n

Convolutional networks, optimization, and visual recognition.

Use with: Backpropagation and automatic differentiation, Optimization, loss, and regularization, Attention and transformers, Generative models and applications.

4. Hugging Face LLM Course

Transformers, attention, and modern model workflows.

Use with: Optimization, loss, and regularization, Convolutional neural networks for vision, Generative models and applications, GPU workflows, scale, and deployment.

Source-Use Standard

Assignments should distinguish among measured notebook evidence, course-provided synthetic evidence, claims supported by these sources, and the student’s own professional judgment. Cite the specific page, section, control, or documentation topic used; a bare homepage link is not adequate evidence.