Resources
Curated directory: courses, readings, and workshop tools.
A curated directory of syllabus-cited courses, papers, and workshop tools — not a second lecture materials index. Jump by category, then open the link you need.
Browse
Courses
4External lecture series and course sites cited in the syllabus.
Neural Networks: Zero to Hero — Andrej Karpathy (opens in new tab)
karpathy.ai
Lecture series on building neural nets and a small GPT from scratch; practical reference for the transformers / small-LLM workshop track.
Sergey Levine — Designing, Visualizing and Understanding Deep Neural Networks (Berkeley CS W182 / 282A) (opens in new tab)
cs182sp21.github.io
Deep learning course site (Spring 2021) with lecture slides, discussions, and homework topics covering nets, CNNs, RNNs, transformers, RL, and generative models — cited for deep-learning foundations in this syllabus.
Zero to Hero notebooks — Andrej Karpathy (opens in new tab)
github.com
Jupyter notebooks that accompany the Zero to Hero video lectures.
Zero to Hero video lectures — Andrej Karpathy (opens in new tab)
youtube.com
Video lectures for the Zero to Hero series.
Browse
Readings
6Core textbooks and papers for classical ML and generative models.
A decoder-only foundation model for time-series forecasting (TimesFM) (opens in new tab)
arxiv.org
Introduces TimesFM, a patched decoder-only transformer foundation model designed for zero-shot time-series forecasting.
An Introduction to Statistical Learning (opens in new tab)
statlearning.com
Primary statistical-learning reference family cited for probability, regression, and classical ML topics (Tibshirani et al.).
Chronos: Learning the Language of Time Series (opens in new tab)
arxiv.org
Presents Chronos, a probabilistic forecasting framework that tokenizes time series and trains transformer language-model architectures for zero-shot prediction.
Denoising Diffusion Probabilistic Models (Ho, Jain, Abbeel) (opens in new tab)
arxiv.org
Canonical DDPM paper (NeurIPS 2020) for the diffusion-models module.
Generative Modeling by Estimating Gradients of the Data Distribution (Song & Ermon) (opens in new tab)
arxiv.org
Score matching with Langevin dynamics / NCSN (NeurIPS 2019) — foundational score-based generative modeling paper by Yang Song et al.
Score-Based Generative Modeling through Stochastic Differential Equations (Song et al.) (opens in new tab)
arxiv.org
Unifies score matching (SMLD) and DDPM under continuous-time SDEs (ICLR 2021); core reading for score-based / diffusion connections.
Browse
Tools
11Language and library docs used in workshops — keep these bookmarked.
Matplotlib (opens in new tab)
matplotlib.org
Core plotting library for visualization workshops.
NetworkX (opens in new tab)
networkx.org
Network analysis toolkit referenced with causality workshops.
NumPy (opens in new tab)
numpy.org
Numerical computing foundation used with Pandas.
Pandas (opens in new tab)
pandas.pydata.org
Tabular data analysis and manipulation.
Plotly (opens in new tab)
plotly.com
Interactive visualization toolkit referenced with causality workshops.
Python (opens in new tab)
docs.python.org
Primary programming language for course workshops.
PyTorch (opens in new tab)
pytorch.org
Deep-learning framework referenced in neural-network workshops.
scikit-learn (opens in new tab)
scikit-learn.org
Classical machine-learning toolkit used with regression and related topics.
Seaborn (opens in new tab)
seaborn.pydata.org
Statistical visualization helpers on top of Matplotlib.
SQL (SQLite documentation) (opens in new tab)
sqlite.org
Relational querying reference for the databases module.
TensorFlow (opens in new tab)
tensorflow.org
Deep-learning framework referenced alongside PyTorch.
Looking for something else?
Lecture slides and notes live on Lectures. Course logistics and policies live on Syllabus. For questions, start with FAQ or Staff.