Fast Neural Kernel Embeddings for General Activations
Explore fast sketching methods for approximating multi-layered NNGP and NTK matrices with various activation functions, overcoming limitations of prior works and achieving significant speedups.
The Quest for Adaptivity in Machine Learning - Comparing Popular Methods
Explore supervised learning algorithms' strengths and weaknesses, comparing k-nearest-neighbor, kernel methods, and neural networks for adaptivity and performance across various problems.
Gauss-Legendre Features for Scalable Gaussian Process Regression
Discover how Gauss-Legendre features enable efficient scaling of Gaussian process regression through low-rank kernel matrix approximation, offering faster computation for large datasets with poly-logarithmic feature requirements.
Gaussian Processes
Explore Gaussian Processes with Daniel Hernández-Lobato, covering key concepts, applications, and advanced techniques in probabilistic machine learning.
A First-Principles Approach to Understanding Deep Learning
Delve into the fundamental principles of deep learning through statistical physics, exploring connections between neural networks, Gaussian processes, and scaling laws with insights from a DeepMind researcher.