Statistical Learning Theory Courses and Certifications

Quantifying Generalization Complexity for Large Language Models

Explore SCYLLA, a framework that measures how well large language models generalize beyond training data by disentangling generalization from memorization across complexity levels, revealing insights about 28 popular LLMs.

• YouTube
• 15 minutes
• Self-Paced
• Free Video

Valerio Velardo - The Sound of AI

Solving Overfitting in Neural Networks

Learn techniques to identify and prevent overfitting in neural networks, including early stopping, audio data augmentation, dropout, and L1/L2 regularization, with practical implementation in a music genre classifier.

• YouTube
• 26 minutes
• Self-Paced
• Free Video

Richard McElreath

Statistical Rethinking 2022 - Overfitting

Explore overfitting in statistical modeling, covering cross-validation, regularization, information criteria, and robust regression techniques for improved predictive accuracy.

• YouTube
• 1 hour 12 minutes
• Self-Paced
• Free Video

Data Science Dojo

VSM, LSA, & SVD - Introduction to Text Analytics with R

Explore advanced text analytics techniques: VSM, LSA, and SVD. Learn to handle high-dimensional data, uncover latent semantic structures, and efficiently map new data in reduced dimensions.

• YouTube
• 38 minutes
• Self-Paced
• Free Video

Yannic Kilcher

Grokking - Generalization Beyond Overfitting on Small Algorithmic Datasets

Explore the fascinating "grokking" phenomenon in neural networks, where sudden pattern recognition leads to perfect generalization. Discover its implications for deep learning and algorithmic datasets.

• YouTube
• 30 minutes
• Self-Paced
• Free Video

Yannic Kilcher

Supervised Contrastive Learning

Explore a novel supervised learning approach that outperforms cross-entropy loss, improving image classification models' performance and robustness across various architectures and data augmentations.

• YouTube
• 30 minutes
• Self-Paced
• Free Video

Yannic Kilcher

Discovering Symbolic Models from Deep Learning with Inductive Biases

Exploring how Graph Neural Networks and symbolic regression can derive accurate symbolic equations from observational data, combining deep learning with physics-inspired inductive biases.

• YouTube
• 46 minutes
• Self-Paced
• Free Video

Institute for Pure & Applied Mathematics (IPAM)

Generalization Theory in Machine Learning

Explore generalization theory in machine learning, covering traditional methods, deep learning, and statistical learning theory. Gain insights into overcoming the curse of dimensionality and understanding learning bounds.

• YouTube
• 1 hour 11 minutes
• Self-Paced
• Free Video

Simons Institute

Distribution Shift as Underspecification, and What We Might Do About It

Explore strategies for addressing distribution shift in deep learning, focusing on underspecification and potential solutions for more robust AI systems.

• YouTube
• 1 hour 1 minute
• Self-Paced
• Free Video

Simons Institute

Implicit Regularization II

Explore implicit regularization in deep learning, focusing on advanced concepts and their impact on model performance and generalization.

• YouTube
• 1 hour 23 minutes
• Self-Paced
• Free Video

Alfredo Canziani

CNN Loss

Explore convolutional neural network loss functions, including mean square error, cross-entropy, and negative log likelihood, with practical examples and implementation insights.

• YouTube
• 42 minutes
• Self-Paced
• Free Video

Data Science Courses

PAC Learnability in Deep Learning - Lecture 20

Delve into statistical machine learning fundamentals, exploring PAC learnability concepts and their application in deep learning, with focus on generalization theories and practical challenges.

• YouTube
• 44 minutes
• Self-Paced
• Free Video

Institute for Advanced Study

Convergence of Nearest Neighbor Classification - Sanjoy Dasgupta

Explore nearest neighbor classification's convergence, consistency, and accuracy rates in various spaces, with insights on smoothness conditions and adaptive classifiers.

• YouTube
• 49 minutes
• Self-Paced
• Free Video

Simons Institute

Implicit Regularization I

Explore implicit regularization in deep learning, covering boosting, complexity control, optimization landscapes, and stochastic gradient descent for effective model training.

• YouTube
• 1 hour 17 minutes
• Self-Paced
• Free Video

Simons Institute

Generalization II

Explore advanced concepts in deep learning, including random averages, losses, and network architectures, with experts from UC Berkeley and MIT.

• YouTube
• 1 hour 23 minutes
• Self-Paced
• Free Video