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 ste…
53:0515K views · 4d 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 · 1w 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!!
24:1852K views · 1mo 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
3:5727K views · 1mo 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
1:01:447.2K 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 · 3mo 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
24:0222K views · 3mo 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 · 4mo 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.
18:278.6K views · 4mo 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...
32:0186K 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 · 9mo 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
5:5934K 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
18:0279K 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
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
Showing the 15 most recent episodes. Everything plays right here on Pod Atlas.
Have thoughts about StatQuest with Josh Starmer? Share them with other listeners.
Discuss this podcast