Statistics, Machine Learning, Data Science, and AI seem like very scary topics, but since each technique is really just a combination of small and simple steps, they are actually quite simple. My goal with StatQuest is to break down the major methodologies into easy-to-understand pieces. That said, I don't dumb down the material. Instead, I build up your understanding so that you are smarter. Contact, Video Index, Etc: https://statquest.org
Watched StatQuest with Josh Starmer? Tell other learners what you think below, or read the discussion around the channel.
53:0519K views · 16d ago · 53:05
Guest: Support Vector Machines
ALSO, here are the files for this livestream We'll spend five days diving into the fundamentals of machine learning, breaking down everything from linear regression and decision trees to random forests and neural networks.
Topics: Awesome song and introduction · Using SVMs in Python · Visualizing the SVC classifier
1:01:0224K views · 21d ago · 1:01:02
Guest: Jon Krohn
If you'd like to support StatQuest, please consider... ...or just donating to StatQuest! paypal venmo: @JoshStarmer
1:15:2034K views · 1mo ago · 1:15:20
Guest: Naive Bayes
We'll spend five days diving into the fundamentals of machine learning, breaking down everything from linear regression and decision trees to random forests and neural networks. Learn more at In this Concepts and Code Livestream, we go over the theory of Naive Bayes and then build a Naive Bayes Classifier in R and Python. DOUBLE BAM!!!
Topics: Classic Naive Bayes (Multinomial) · Gaussian Naive Bayes · Questions! · Coding a Naive Bayes Classifier with AI · A special silly song for Scotland · More questions!!
24:1853K views · 2mo ago · 24:18
There's a new activation function in town. Actually, there are a bunch of new activation functions in town. This StatQuest gives you the essential information you need at the very start, followed by a surprisingly interesting derivation of each of these new activation functions. It turns out that Dropout is all you need. Who knew?
Topics: Awesome song and introduction · TL;DW (too long, didn't watch) · Background and why the Sigmoid and ReLU activation functions fail · How dropout improves training · GELU = Input dependent dropout · General equation for a class of activation functions
3:5726K views · 2mo ago · 3:57
Since September 4th is Global Frank Starmer Day (and also his birthday), we'll celebrate by talking about another three (3) more lessons from him that influenced me over the span of my career so far. If you'd like to support StatQuest, please consider...
Topics: Awesome song and introduction · Always be curious · Exercise your mind and your body · Make friends with everyone you can
1:01:447K views · 2mo ago · 1:01:44
Guest: Brandon Rohrer
Made with Restream Studio. Livestream on 30+ platforms at once via Feel free to post your questions beforehand or ask them live!
26:3818K views · 4mo ago · 26:38
In the StatQuest, Optimization with Linear Programming (and the Simplex Algorithm), Main Ideas!!!, we learned the Main Ideas of Linear Programming and the Simplex Algorithm. Now let's dive into the nitty gritty details of how it's done, one step at a time. BAM!
Topics: Awesome song and introduction · Terminology and setting up the matrix for the first example · Solving the first example with the Simplex Algorithm · Setting up the matrix for the second example · Solving the second example with the Simplex Algorithm
24:0223K views · 4mo ago · 24:02
Guest: Linear Programming
If you've ever wondered how airplane schedules are optimized, warehouses are optimized or how crops are optimized, then you've wondered about Linear Programming. This StatQuest introduces you to the main ideas of how Linear Programming can be used to optimize problems, and, specifically, we'll learn about the Simplex Algorithm. BAM!
Topics: Awesome song and introduction · An example of an optimization problem · Why we need Linear Programming · The main ideas of the Simplex Algorithm · A fancier example
10:4815K views · 5mo ago · 10:48
NOTE: This StatQuest is the updated version of the original Random Forests Part 3 and includes two minor corrections. Last time we talked about how to create, use and evaluate random forests. Now it's time to see how they can deal with missing data and how they can be used to cluster samples, even when the data comes from all kinds of crazy sources.
18:279K views · 5mo ago · 18:27
One of the best ways to prevent p-hacking is to adjust p-values for multiple testing. This StatQuest explains how the Benjamini-Hochberg method corrects for multiple-testing and FDR. If you'd like to support StatQuest, please consider...
32:0194K views · 5mo ago · 32:01
This StatQuest takes you through basic linear regression, step-by-step, explaining both the method and the motivation behind the method. Each step is broken down into easy-to-understand parts so that anyone can develop a deep understanding of how to do and interpret a linear regression. BAM!
Topics: Awesome song and introduction · Quantifying Prediction Accuracy with the Sum of Squared Residuals · Fitting a Line with Least Squares · Assessing Predictions with R-Squared · Assessing Predictions with a p-value · Putting it all together
22:5141K views · 10mo ago · 22:51
In this StatQuest we explain how AI, like ChatGPT, works. However, instead of using lots of complicated math, we're going to use a super simple example and super simple terms. If math scares you, this StatQuest is for you! And even if math doesn't scare you, this StatQuest will give you a deeper understanding of what goes on under the hood. BAM!
Topics: Awesome song and introduction · Main Idea #1: How an AI writes poems · Main Idea #2: How an AI is created · Main Idea #3: How an AI is aligned to do useful things
5:5933K views · 1y ago · 5:59
This year, for my Pop's birthday, we're going to hear from current tech industry leaders that all started out working for my pop in the IT-Lab at the Medical University of South Carolina. These leaders have started all kinds of startups, pioneered Oscar award winning computer graphics and designed the best user interfaces that I've ever seen. I believe the advice that these leaders give is so valuable that I've awarded them the rare, but highly coveted, Quadruple Bam!!!
Topics: Awesome song and introduction · Ask questions to uncover the real problem · Value everyone's contribution · Small steps lead to significant progress · Unix pipes as a software design pattern
18:0282K views · 1y ago · 18:02
Guest: Human Feedback
Generative Large Language Models, like ChatGPT and DeepSeek, are trained on massive text based datasets, like the entire Wikipedia. However, this training alone fails to teach the models how to generate polite and useful responses to your prompts. Thus, LLMs rely on Supervised Fine-Tuning and Reinforcement Learning with Human Feedback (RLHF) to align the models to how we actually want to use them.
Topics: Awesome song and introduction · Pre-Training an LLM · Supervised Fine-Tuning · Reinforcement Learning with Human Feedback (RLHF) · RLHF - training the reward model · RLHF - using the reward model
📄 Mentioned papers: NOTE: This video is based on the original manuscript for Instruct-GPT
25:0134K views · 1y ago · 25:01
Guest: Neural Networks
Here we go through the math required to update a parameter in a neural network using reinforcement learning and we do it one step a time. We show how the derivatives are calculated (BAM!), then updated (DOUBLE BAM!!), and then used to optimize the parameters (TRIPLE BAM!!!).
Topics: Awesome song and introduction · Calculating a derivative · Updating the derivative with a reward · Updating a parameter in the neural network · A second example
24:0062K views · 1y ago · 24:00
Guest: Neural Networks
Reinforcement Learning has helped train neural networks to win games, drive cars and even get ChatGPT to sound more human when it responds to your prompt. This StatQuest covers the essential concepts of how this process works. BAM!
Topics: Awesome song and introduction · Backpropagation review · The problem with standard backpropagation · Taking a guess to calculate the derivative · Using a reward to update the derivative · Alternative rewards
18:13120K views · 1y ago · 18:13
Guest: Reinforcement Learning
Reinforcement Learning is one of the most useful methodologies for training AI systems right now, and, while it might seem intimidating, is actually very similar to how you and I learn. This StatQuest cuts through the hype and helps you learn the essential concepts behind reinforcement learning one step at a time.
Topics: Awesome song and introduction · Updating the Policy, part 1 · Understanding the Learning Rate · Updating the Policy, part 2 · Reinforcement Learning Terminology
59:559K views · 1y ago · 59:55
If you'd like to support StatQuest, please consider... ...or just donating to StatQuest! paypal venmo: @JoshStarmer
18:52125K views · 1y ago · 18:52
Encoder-Only Transformers are the backbone for RAG (retrieval augmented generation), sentiment analysis and classification problems, and clustering. This StatQuest covers the main ideas of how these powerhouses do what they do so well, making sure each step is clearly explained!
Topics: Awesome song and introduction · Word Embedding · Positional Encoding · Attention · Applications of Encoder-Only Transformers · RAG (Retrieval-Augmented Generation)
▶️ Mentioned videos: Transformers · Decoder-Only Transformers · The Matrix Math Behind Transformers · Coding a Decoder-Only Transformer Scratch in PyTorch
54:355K views · 2y ago · 54:35
Join me, Luis Serrano for a Q&A Livestream where we answer your questions about AI, ML, Statistics and all things Data Science!!! TRIPLE BAM!!! If you'd like to support StatQuest, please consider... ...or just donating to StatQuest! paypal venmo: @JoshStarmer
24:089K views · 2y ago · 24:08
Guest: Nana Janashia
If you'd like to support StatQuest, please consider... ...or just donating to StatQuest!
4:408K views · 2y ago · 4:40
Since September 4th is Global Frank Starmer Day (and also his birthday), we'll celebrate by talking about a few more lessons from him that influenced me over the span of my career so far. If you'd like to support StatQuest, please consider...
54:497K views · 2y ago · 54:49
Guest: Abbas Merchant
If you'd like to support StatQuest, please consider... ...or just donating to StatQuest!
59:566K views · 2y ago · 59:56
Join me, Luis Serrano and for a Q&A Livestream where we answer your questions about AI, ML, Statistics and all things Data Science!!! TRIPLE BAM!!! If you'd like to support StatQuest, please consider... ...or just donating to StatQuest! paypal venmo: @JoshStarmer
31:11131K views · 2y ago · 31:11
In this StatQuest we walk through the code required to code your own ChatGPT like Transformer in PyTorch and we do it one step at a time, with every little detail clearly explained. NOTE: This StatQuest assumes that you are already familiar with the concepts behind... Decoder-Only Transformers The Essential Matrix Algebra for Neural Networks The Matrix Math Behind Transformers
Topics: Awesome song and introduction · Loading the modules · Creating the training dataset · Coding Position Encoding · Coding Attention · Coding a Decoder-Only Transformer
▶️ Mentioned videos: Decoder-Only Transformers · The Essential Matrix Algebra for Neural Networks · The Matrix Math Behind Transformers · The full Neural Networks playlist, the basics to AI, is
28:256K views · 2y ago · 28:25
Guest: Dr. Amy Finnegan
In this episode we have special guest Dr. Amy Finnegan, the Deputy Director of Data Science at IntraHealth International. Amy is a demographer and data scientist with over 10 years of experience working in global health, development, and data science in emerging economies on four continents. Amy is also an adjunct faculty at Duke University’s Global Health Institute, and before these jobs, she was a research scholar at Duke.
32:335K views · 2y ago · 32:33
Guest: Xavier Godoy
In this episode we have special guest Xavier Godoy, the director of customer experience and automation at HBX Group in Mallorca, Spain. Xavier has had a career driven by curiosity and a desire to learn more while simultaneously making sure that customer satisfaction is always is the focus of his efforts.
36:555K views · 2y ago · 36:55
Guest: Tommy Tang
In this episode we have special guest Tommy Tang, the Director of Computational Biology at Immunitas Therapeutics. Tommy is a computational biologist with over ten years of computational experience and six years' wet lab experience committed to reproducible research and open science. At Immunitas Therapeutics, Tommy employs a single-cell sequencing platform to dissect the biology of immune cells in human tumors by using machine learning approaches and cloud computing.
57:306K views · 2y ago · 57:30
Join me and Luis Serrano for a Q&A Livestream where we answer your questions about AI, ML, Statistics and all things Data Science!!! TRIPLE BAM!!! If you'd like to support StatQuest, please consider... ...or just donating to StatQuest! paypal venmo: @JoshStarmer
37:275K views · 2y ago · 37:27
Guest: Simon Stochhom
In this episode we have special guest Simon Stochhom, a lecturer at UCL in Denmark. Simon applies machine learning, especially deep learning, to images, video and time series in wide variety of settings. And by “wide variety”, I really mean it. Simon is fearless when it comes to seizing opportunities that come up and somehow turns them all into success stories.
3:218K views · 2y ago · 3:21
Guest: Log_e Song
When I first started making StatQuest videos it never dawned on me that people would try to re-do my math on their own. I was also new to explaining things and just assumed that everyone already knew that, in statistics and machine learning, when you use the log, you use base 'e'. Big rookie mistake! Ever since then, people have been re-doing my math using log base 10 and posting comments about how their math is not working out.
🎵 Mentioned music: the track on Spotify
35:498K views · 2y ago · 35:49
Guest: Brian Risk
In this episode we have special guest Brian Risk, a multi-talented data scientist and the President and Founder of devra.ai, a company specializing in automated coding. Brian is also a great personal friend of mine and an amazing musician.
23:4397K views · 2y ago · 23:43
Transformers, the neural network architecture behind ChatGPT, do a lot of math. However, this math can be done quickly using matrix math because GPUs are optimized for it. Matrix math is also used when we code neural networks, so learning how ChatGPT does it will help you code your own. Thus, in this video, we go through the math one step at a time and explain what each step does so that you can use it on your own with confidence.
Topics: Awesome song and introduction · Word Embedding · Position Encoding · Self Attention · Residual Connections · Decoder Word Embedding and Position Encoding
▶️ Mentioned videos: Transformers · The essential matrix algebra for neural networks
35:315K views · 2y ago · 35:31
Guest: Fabio Urbina
In this episode we have special guest Fabio Urbina, an Associate Director at Collaborations Pharmaceuticals. Fabio combines computational tools and machine learning with classical small-molecule, molecular, and cell biology techniques to address previously-difficult to probe scientific problems. Specifically, Fabio finds solutions to drug discovery with machine learning.
27:1310K views · 2y ago · 27:13
Guest: Kushi Jain
In this episode we have special guest Kushi Jain, who works in Data Analytics Development at John Deere and in the Master’s in Computer Science - Data Science at the University of Illinois. Having recently graduated with her bachelor’s, Kushi participated in the data science club and also completed several internships at John Deere.
33:129K views · 2y ago · 33:12
Guest: Achal Dixit
In this episode we have special guest Achal Dixit, a Data Scientist at Delhivery, the largest fully integrated logistics services in India. Achal solves problems using Data, statistics, and machine learning with a focus on business and people. Before Delhivery, Achal was a Business Technology Analyst at ZS. And before that, Achal was a research assistant at Imperial College London.
31:1320K views · 2y ago · 31:13
Guest: Rick Marks
In this episode we have special guest Rick Marks, a professor at the University of North Carolina Chapel Hill School of Data Science and Society. Before UNC, Rick was a director at Google's Advanced Technology and Projects group, exploring new interaction approaches for ambient computing environments. And before that, Rick founded the PlayStation Magic Lab at PlayStation R&D.
30:0187K views · 2y ago · 30:01
Although you don't need to know matrix algebra to understand the ideas behind neural networks, if you want to code them or read the latest manuscripts about the field, then you'll need to understand matrix algebra.
Topics: Awesome song and introduction · Linear transformations in matrix notation · Matrix multiplication · Matrix multiplication consolidates a sequence of linear transformations · Transposing a matrix · Matrix notation and equations
▶️ Mentioned videos: Video on youtube.com · Video on youtube.com · Video on youtube.com · Video on youtube.com
32:0254K views · 3y ago · 32:02
Word embedding is the first step in lots of neural networks, including Transformers (like ChatGPT) and other state of the art models. Here we learn how to code a stand alone word embedding network from scratch and with nn.Linear. We then learn how to load and use pre-trained word embedding values with nn.Embedding.
Topics: Awesome song and introduction · Importing modules · Encoding the training data · Word Embedding from scratch · Graphing the embedding values · Printing out predicted words
▶️ Mentioned videos: Video on youtube.com
2:3027K views · 3y ago · 2:30
The Golden Play Button is usually super confusing. In this video, we break it down and walk you through it one-step-at-a-time. By the end of this StatQuest, you'll completely understand The Golden Play Button. If you'd like to support StatQuest, please consider...
6:4612K views · 3y ago · 6:46
Since September 4th is Global Frank Starmer Day (and also his birthday), I thought we'd celebrate by talking about another three lessons that influenced my older sister and my older brothers. If you'd like to support StatQuest, please consider...
Topics: Awesome song and introduction · To solve problems, draw a picture · To remember, draw a picture · To solve a problem, start.
36:45249K views · 3y ago · 36:45
Transformers are taking over AI right now, and quite possibly their most famous use is in ChatGPT. ChatGPT uses a specific type of Transformer called a Decoder-Only Transformer, and this StatQuest shows you how they work, one step at a time. And at the end (at 32:14), we talk about the differences between a Normal Transformer and a Decoder-Only Transformer. BAM!
Topics: Awesome song and introduction · Word Embedding · Position Encoding · Masked Self-Attention, an Autoregressive method · Residual Connections · Generating the next word in the prompt
▶️ Mentioned videos: The Chain Rule · Gradient Descent · Backpropagation Main Ideas · Backpropagation Details Part 1
36:151.2M views · 3y ago · 36:15
Transformer Neural Networks are the heart of pretty much everything exciting in AI right now. ChatGPT, Google Translate and many other cool things, are based on Transformers. This StatQuest cuts through all the hype and shows you how a Transformer works, one-step-at-a time.
Topics: Awesome song and introduction · Word Embedding · Positional Encoding · Self-Attention · Encoder and Decoder defined · Decoder Word Embedding
▶️ Mentioned videos: The Chain Rule · Gradient Descent · Backpropagation Main Ideas · Backpropagation Details Part 1
15:51482K views · 3y ago · 15:51
Attention is one of the most important concepts behind Transformers and Large Language Models, like ChatGPT. However, it's not that complicated. In this StatQuest, we add Attention to a basic Sequence-to-Sequence (Seq2Seq or Encoder-Decoder) model and walk through how it works and is calculated, one step at a time. BAM!!!
Topics: Awesome song and introduction · The Main Idea of Attention · A worked out example of Attention · The Dot Product Similarity · Using similarity scores to calculate Attention values · Using Attention values to predict an output word
▶️ Mentioned videos: Video on youtube.com
📄 Mentioned papers: arXiv 1409.0473 · arXiv 1508.04025
Showing the 44 most recent videos. Everything plays right here on The Pod Atlas.
Have thoughts about StatQuest with Josh Starmer? Share them with other learners.
Write a review← Back to all educational channels