Online Channels

These are YouTube channels that explain the course’s material in another voice, go deeper than the notes, or show where the field is heading. The lists hold 93 entries from 76 channels; a channel useful for two topics appears under both.

The lists are curated for this course, not drawn from a ranking. Each entry’s level and description are a judgment of the channel’s content, and the lists are updated as channels change. The description of each channel links to specific playlists or videos on the table’s topic, so a channel listed under two topics links to different material in each. We last confirmed that every link opens the channel, playlist or video it names on 2026-10-09.

Each channel has a level:

• Intro: no prerequisites beyond high-school mathematics.
• Intermediate: upper-division or graduate lecture series.
• Advanced: assumes a graduate background.
• Research: talks and paper reviews.
• General: for a popular audience; good for motivation, not for study.

1 Machine learning

Table 1: Machine learning channels, with level and what each is good for.

Channel	Level	Good for
StatQuest with Josh Starmer	Intro	Short, step-by-step explanations of nearly every method in DATA 571 (trees, boosting, SVMs, PCA, regularization).
Luis Serrano Academy	Intro	Friendly visual intuition for neural nets, transformers and classical ML.
3Blue1Brown	Intro	The “Neural networks” series: backpropagation, gradient descent and transformers drawn from first principles.
Welch Labs	Intro	Carefully animated stories of ML ideas (learning to see, the perceptron, backpropagation).
Stanford Online	Intermediate	Full lecture series: CS229 (machine learning), CS231n (vision), CS224N (NLP), and Hastie and Tibshirani’s Statistical Learning (ISLR).
caltech	Intermediate	Yaser Abu-Mostafa’s “Learning from Data” (CS 156): the theory of generalization, VC dimension and bias-variance, very clear.
Kilian Weinberger	Intermediate	Cornell CS4780 lectures; close in scope and level to DATA 571.
MIT OpenCourseWare	Intermediate	Lectures such as Introduction to Machine Learning and Introduction to Neural Networks and Deep Learning, plus related ML and statistics courses.
Alfredo Canziani	Intermediate	NYU Deep Learning (with Yann LeCun): history of deep learning, energy-based models, PyTorch, graduate level.
mathematicalmonk	Intermediate	Older but rigorous short lectures on ML and probability from a mathematical viewpoint.
Steve Brunton	Intermediate	SVD, PCA, regression and ML for science and engineering; pairs with his “Data-Driven Science and Engineering” book.
Mutual Information	Intermediate	Polished explainers on the theory behind ML: bias-variance, information theory, RL.
Andrej Karpathy	Advanced	“Neural Networks: Zero to Hero”: building autograd, language models and GPT from scratch in Python.
Pieter Abbeel	Advanced	Berkeley courses on his courses page: Deep Unsupervised Learning (Spring 2024, the source of the semi-supervised lecture video; also Spring 2020), Foundations of Deep RL and CS287 Advanced Robotics (Fall 2019), for deep learning and reinforcement learning courses after this one.
Umar Jamil	Advanced	Line-by-line implementations of transformers, diffusion models and LLM techniques such as LoRA in PyTorch.
DeepLearningAI	Intermediate	Andrew Ng’s Heroes of Deep Learning interviews and lessons from DeepLearning.AI courses, including the Machine Learning Specialization and the Deep Learning Specialization’s neural networks, regularization and optimization, project structuring and convolutional network courses.
Jeremy Howard	Intermediate	fast.ai “Practical Deep Learning for Coders”: top-down, code-first.
sentdex	Intro	Practical Python ML and deep learning tutorials.
Yannic Kilcher	Research	Long walk-throughs of ML papers, including classic papers.
AI Coffee Break with Letitia	Research	Short, accurate explainers on transformers, LLMs and diffusion models.
Two Minute Papers	General	Quick tours of new research results in AI and deep learning, plus graphics and generative models.
Machine Learning Street Talk	Research	Long interviews with ML researchers on theory and open problems, such as Francois Chollet on intelligence and generalisation and Andrew Wilson on why huge models work.
Simons Institute for the Theory of Computing	Research	Workshop talks on learning theory, optimization and deep learning theory.

2 Statistics

Table 2: Statistics channels, with level and what each is good for.

Channel	Level	Good for
StatQuest with Josh Starmer	Intro	Hypothesis tests, p-values, linear models, maximum likelihood, explained slowly and visually.
jbstatistics	Intro	Concise, accurate introductory probability and inference: distributions, sampling, confidence intervals.
5MinuteAI	Intro	Five-minute videos that follow the University of Washington’s CSE 312: Probability & Statistics runs from sets and counting through random variables, limit theorems, maximum likelihood and hypothesis testing.
Khan Academy	Intro	AP and college intro statistics and probability, with exercises on its site.
CrashCourse	General	The “Statistics” series: a fast conceptual overview.
zedstatistics	Intro	Regression, time series and econometrics with a practical bent.
Brandon Foltz	Intro	Patient walk-throughs of regression, ANOVA and logistic regression.
Very Normal	Intermediate	Biostatistics-flavoured explainers on inference, Bayesian ideas and study design.
ritvikmath	Intermediate	Data-science statistics: time series, Bayesian methods, MCMC, with whiteboard derivations.
Ben Lambert	Intermediate	Bayesian statistics and econometrics; companion to “A Student’s Guide to Bayesian Statistics”.
Richard McElreath	Intermediate	“Statistical Rethinking” lectures: Bayesian modelling and causal inference with DAGs.
Harvard University	Intermediate	Joe Blitzstein’s Stat 110 (probability) lectures, the standard reference course.
MIT OpenCourseWare	Intermediate	18.650 (statistics for applications) and 18.05 (probability and statistics).
Stanford Online	Intermediate	Hastie and Tibshirani’s “Statistical Learning”, the lectures for ISLR.
Mutual Information	Intermediate	Statistical and information-theoretic foundations of ML.
Julia Silge	Intermediate	Screencasts of modelling real data with tidymodels in R, such as lasso regression.
Posit PBC	Intermediate	Talks and tutorials on R, Quarto, Shiny and the tidyverse.

3 Artificial intelligence

Methods are under Machine learning; this list covers large language models, reinforcement learning, AI safety, and the research field.

Table 3: AI channels, with level and what each is good for.

Channel	Level	Good for
3Blue1Brown	Intro	Visual explanations of how transformers and LLMs work, from attention to next-token prediction.
Andrej Karpathy	Intro	“Intro to Large Language Models” and “Deep Dive into LLMs” for anyone; “Let’s build GPT” for coders.
Google DeepMind	Intermediate	David Silver’s reinforcement learning course, the UCL x DeepMind lectures, and research talks.
Stanford Online	Intermediate	CS221 (AI principles), CS224N (NLP with deep learning), CS234 (reinforcement learning).
MIT OpenCourseWare	Intermediate	Patrick Winston’s 6.034 Artificial Intelligence, a classic survey of the field.
Hugging Face	Intermediate	Courses and tutorials on transformers, fine-tuning and open models.
Umar Jamil	Advanced	Coding LLM components (attention, LoRA, RLHF) from scratch.
Robert Miles AI Safety	General	Clear introductions to AI alignment problems: specification gaming, reward hacking, mesa-optimization.
Anthropic	General	An AI lab’s talks and interviews on interpretability, alignment and AI policy.
AI Explained	General	Careful coverage of new model releases and benchmarks, reading the papers rather than the press.
Yannic Kilcher	Research	Paper walk-throughs, including the landmark LLM and RL papers: Attention Is All You Need, GPT-3 and learning to summarize from human feedback.
Machine Learning Street Talk	Research	In-depth interviews on reasoning, generalization and AI risk.
Dwarkesh Patel	General	Long, well-prepared interviews with AI researchers and lab leaders, such as Ilya Sutskever, Dario Amodei and Demis Hassabis.
Lex Fridman	General	Long-form interviews; the early episodes and his MIT deep learning lectures (basics, state of the art) are the AI-relevant part.
Artem Kirsanov	Intermediate	Computational neuroscience and its links to AI (Hopfield networks, predictive coding); see his AI and machine learning playlist.

4 Computer science

Table 4: Computer science channels, with level and what each is good for.

Channel	Level	Good for
CS50	Intro	Harvard’s CS50 lectures, including CS50 Python and CS50 AI.
MIT OpenCourseWare	Intermediate	6.006 (algorithms), 6.046 (design and analysis of algorithms), 6.042 (math for CS).
Missing Semester	Intro	MIT’s “Missing Semester”: the shell, git, editors, debugging; useful before any data course.
freeCodeCamp.org	Intro	Full-length courses on Python, SQL, git, data structures and ML tooling.
Computerphile	General	Short interviews with CS researchers on algorithms, security, AI and theory.
Abdul Bari	Intermediate	Algorithms by hand on a whiteboard: complexity, dynamic programming, graphs.
WilliamFiset	Intermediate	Data structures and graph algorithms with animations and code.
Reducible	Intro	Animated explanations of algorithms such as the FFT, compression and PageRank.
Spanning Tree	Intro	Animated explainers on algorithms and how computers work.
Ben Eater	Intermediate	Building an 8-bit computer and networking from the hardware up.
Core Dumped	Intro	Operating systems and memory explained with animations.
Low Level	Intermediate	C, memory safety and systems security.
CMU Database Group	Intermediate	Andy Pavlo’s database systems courses (15-445/645, 15-721).
Sebastian Lague	General	Coding adventures in simulation, graphics and procedural generation; good for motivation.
NeetCode	Intro	Algorithm interview problems solved and explained in Python.
Fireship	General	Fast overviews of languages and tools (“in 100 seconds”).
Simons Institute for the Theory of Computing	Research	Theoretical CS workshops: complexity, cryptography, learning theory.

5 Mathematics

Table 5: Mathematics channels, with level and what each is good for.

Channel	Level	Good for
3Blue1Brown	Intro	“Essence of linear algebra” and “Essence of calculus”: the geometric intuition behind DATA 571’s math.
Khan Academy	Intro	Calculus, linear algebra and multivariable calculus at an introductory pace.
Professor Leonard	Intro	Full-length calculus 1, calculus 2 and statistics lectures, very thorough.
Dr. Trefor Bazett	Intro	Calculus, linear algebra, discrete math and probability, with visual explanations.
MIT OpenCourseWare	Intermediate	Gilbert Strang’s 18.06 (linear algebra) and 18.065 (matrix methods in data analysis and ML).
Oxford Mathematics	Intermediate	Undergraduate student lectures from Oxford, including probability, plus public lectures.
The Bright Side of Mathematics	Intermediate	Rigorous short series on linear algebra, real analysis, measure theory and probability theory.
Steve Brunton	Intermediate	SVD, Fourier analysis, differential equations and optimization for data science.
Michael Penn	Intermediate	Number theory, abstract algebra and problem solving at a blackboard.
Professor Macauley	Intermediate	Visual group theory and advanced linear algebra lectures (Clemson).
Richard E Borcherds	Advanced	A Fields medallist’s lecture courses on algebra, geometry and number theory.
Mathemaniac	Intro	Visual explainers on complex analysis, group theory and topology, such as the Euler characteristic.
Aleph 0	Intro	Short, elegant explanations of abstract algebra and analysis ideas.
Mathologer	General	Deep dives into beautiful proofs.
Numberphile	General	Interviews with mathematicians, such as Grant Sanderson of 3Blue1Brown, on accessible problems like matrix factorization.
Stand-up Maths	General	Matt Parker’s playful investigations of maths in the wild, such as MENACE, a matchbox machine that learns and Runge spikes.
Zach Star	General	How mathematics is used in engineering and science, such as eigenvectors and eigenvalues.
Institute for Advanced Study	Research	Seminars and lectures from IAS, including the Theoretical Machine Learning Lecture Series and a workshop on optimization, statistics and machine learning.
Isaac Newton Institute for Mathematical Sciences	Research	Programme talks, plus organiser interviews on statistical scalability and prediction uncertainty in machine learning.
Institute for Pure & Applied Mathematics (IPAM)	Research	UCLA’s applied-math workshops, many on ML, optimization and data science.
Fields Institute	Research	Lectures and workshops in pure and applied mathematics, including the Mathematical AI Seminar and the Mathematical Foundations of AI graduate course.

6 Departments and programs

These are academic units, not channels. Their course listings, seminars and research pages extend the notes beyond this course.

Table 6: Departments and programs, with what each is good for.

Unit	Good for
UC Berkeley College of Computing, Data Science, and Society	Berkeley’s college for computing, data science and their societal context; its pages list the college’s programs, courses and research.
UC Berkeley Department of Statistics	The department’s courses, seminars and research relevant to statistical learning.