Machine Learning Basics: A Speedrun - IPAM at UCLA
Speedrun through machine learning fundamentals, covering key concepts like inductive bias, regularization, and kernel methods for both linear and nonlinear regression.
Factorization of Completely Positive Matrices Using Iterative Projected Gradient Steps
Explore iterative projected gradient steps for factorizing completely positive matrices in this advanced mathematical lecture from the University of Vienna.
Learning Deep Matrix Factorizations Via Gradient Descent - Implicit Bias Towards Low Rank
Explores implicit bias in gradient descent algorithms for deep matrix factorizations, analyzing convergence to low-rank matrices and discussing implications for tensor decompositions.
Building a Winning Deep Learning Recommender System - Grandmaster Series Episode 5
Learn how Kaggle Grandmasters built a winning Deep Learning Recommender System for Booking.com, using matrix factorization, ensemble models, and data augmentation techniques. Gain insights into GPU-accelerated recommender systems.
Federated Learning with Formal User-Level Differential Privacy Guarantees
Explore federated learning with differential privacy guarantees, covering key concepts, challenges, and algorithms like DP-SGD and DP-FTRL, with insights on privacy-utility trade-offs and future directions.
Empirical Bayes Matrix Factorization and Parts-Based Representations
Explore empirical Bayes matrix factorization and parts-based representations in computational genomics, focusing on tumor transcriptional heterogeneity analysis from single-cell RNA-seq data.
Hierarchical Interpolative Factorization
Explore efficient matrix factorization techniques for differential and integral operators, focusing on interpolative decomposition and its applications in various dimensions and problem types.
Factorization Methods in Structure from Motion (SFM)
Explore factorization techniques in Structure from Motion (SFM) to enhance 3D reconstruction and computer vision skills.
How Does Netflix Recommend Movies? Matrix Factorization
Explore matrix factorization in recommender systems, focusing on Netflix's movie recommendation algorithm. Learn about dependencies, error functions, and predicting user ratings.
Non-convex Matrix Sensing: Breaking the Quadratic Rank Barrier in Sample Complexity
Explore groundbreaking research on non-convex matrix sensing, focusing on improved algorithms for reconstructing low-rank matrices using fewer samples through innovative probabilistic decoupling methods.
Building Decision-Making Assistants: Introduction to Recommendation Systems
Explore recommendation systems from basic classification to advanced Netflix-style matrix factorization, covering content filtering, collaboration models, SVD, and PCA techniques.
Discover BigQuery ML - Building Machine Learning Models with SQL
Explore BigQuery ML to build, train, and deploy machine learning models using SQL for marketing use cases like customer segmentation and purchase prediction.
The Geometry of the Deep Linear Network
Explore the geometric principles underlying deep linear networks, uncovering their mathematical foundations and implications for machine learning algorithms.
The Geometry of the Deep Linear Network - Part 2
Explore the intricate geometry of deep linear networks, delving into advanced mathematical concepts and their applications in machine learning architectures.
Supervised Quantile Normalization for Matrix Factorization Using Optimal Transport
Explore advanced matrix factorization techniques using optimal transport for quantile normalization. Learn applications in dimensionality reduction and genomics from expert Marco Cuturi.
The Exponential Type Conjecture for Quantum Connections
Delve into quantum connections and formal singularity types, exploring the exponential type conjecture proof for closed monotone symplectic manifolds through mod p arguments and Katz's theorem.
Dimensionality Reduction: Rank-k Approximation and Eigen-decomposition - Lecture 20
Dive into advanced dimensionality reduction techniques, exploring rank-k approximation, eigen-decomposition, and power method applications in data analysis.