Beyond Lazy Training for Over-parameterized Tensor Decomposition
Explore tensor decomposition in over-parameterized settings, analyzing gradient descent variants and their performance beyond lazy training regimes for improved optimization in neural networks.
Differential Calculus on Persistence Barcodes
Explore differential calculus on persistence barcodes, covering differentiability concepts, lifts to ordered barcodes, and applications in topological data analysis.
Math for Machine Learning - Exercises: Probability
Hands-on exercises exploring probability concepts in machine learning, including entropy, cross-entropy, and Gaussian distributions, with practical applications and insights.
On Quantum Linear Algebra for Machine Learning - IPAM at UCLA
Explore quantum singular value transformation for machine learning, its applications, limitations, and connections to classical algorithms in this technical talk on quantum linear algebra.
Introductory Lectures on First-Order Convex Optimization - Lecture 1
Explore first-order convex optimization, covering gradient descent, complexity, smoothness, and Nesterov's accelerated algorithm. Gain insights into theoretical foundations and practical applications in machine learning.
Beyond Lazy Training for Over-parameterized Tensor Decomposition
Explore over-parameterization in tensor decomposition, comparing lazy training and gradient descent approaches. Discover how gradient descent can leverage low-rank data structure beyond lazy training regimes.
Can Learning Theory Resist Deep Learning? - Francis Bach, INRIA
Exploring global convergence of gradient descent in non-convex optimization for deep learning, highlighting challenges in bridging theory and practice in machine learning algorithms.
Optimization: Higher-order Methods - Part 1
Explore advanced optimization techniques using higher-order methods, focusing on theoretical foundations and practical applications in algorithm design and data structures.
Optimization: First-order Methods - Part 2
Explore advanced first-order optimization techniques, focusing on accelerated gradient methods and their applications in solving complex algorithmic problems.
Optimization: First-order Methods - Part 1
Explore first-order optimization methods with Alina Ene, focusing on fundamental techniques and algorithms for efficient problem-solving in data structures and algorithm design.
Geometry and Topology of Spaces of Structured Matrices
Explore geometric and topological aspects of structured matrices, including momentum maps, frame spaces, and applications in compressed sensing and graph theory.
Maximum Likelihood Estimate for Logistic Regression
Explore maximum likelihood estimation for logistic regression, covering likelihood functions and gradient descent techniques.
Mathematics for Deep Neural Networks: Energy Landscape and Open Problems - Lecture 5
Explore the energy landscape of deep neural networks and future challenges in statistical theory, focusing on gradient descent methods and important developments.
Introduction to the Dynamics of Disordered Systems - Equilibrium and Gradient Descent - Part 3
Explore equilibrium dynamics and gradient descent in disordered systems, focusing on theoretical concepts and practical applications in statistical physics and optimization.
A Primer in Machine Learning
Explore machine learning fundamentals, including hypotheses, data representation, algorithms like regression and classification, and practical implementation using MATLAB.
Towards Collaborative Decentralized Learning
Explore cutting-edge decentralized learning concepts, from federated systems to Byzantine-robust training, with insights into collaborative and personalized optimization approaches.
Neurosymbolic Programming - An Emerging Subarea of AI
Explore neurosymbolic programming's fusion of deep learning and program synthesis, discovering how it enhances AI robustness, interpretability, and efficiency through symbolic and neural approaches.
Building and Optimization of Neural Networks from Scratch with NumPy
Master neural network development from ground up using NumPy, covering architecture design, propagation methods, and optimization techniques for building efficient deep learning models without frameworks.
Machine Learning Mid-Semester Review - Lecture 15
Dive into a comprehensive mid-semester review of fundamental machine learning concepts, covering key principles and techniques explored throughout the academic term.
Machine Learning: Neural Networks and Deep Learning - Lecture 18
Dive into advanced data science concepts through comprehensive coverage of machine learning principles, algorithms, and practical applications in this university-level lecture.
Machine Learning Fundamentals - Lecture 16
Dive into advanced data science concepts through comprehensive coverage of machine learning principles, algorithms, and practical applications in this university-level lecture.