Lecture

Neural networks under SGD

In course
DEMO: enim adipisicing
Nisi adipisicing commodo voluptate nostrud. Commodo quis culpa amet veniam incididunt mollit. Elit sint ut ullamco fugiat aliqua id excepteur eiusmod ut laborum. Fugiat eiusmod officia nulla eiusmod. Culpa elit amet irure qui consectetur ut ipsum sunt mollit culpa id enim est.
Login to see this section
Description

This lecture covers the concept of neural networks trained using Stochastic Gradient Descent (SGD). It explains the relationship between the number of neurons in hidden layers and the number of samples, as well as the parameters involved in the network. The lecture delves into the square loss function and the process of taking small steps to optimize the network. It also discusses the evolution of particles in the network, the interpretation of true risk, and the implications of replacing particles with density. The lecture concludes with an exploration of limits and justifications in the context of neural network training.

Instructors (2)
esse et proident eiusmod
Amet magna duis culpa cillum quis velit aute qui velit. Dolore consectetur quis est dolore dolor eiusmod Lorem laboris sint proident. Adipisicing proident irure consectetur dolor voluptate voluptate duis eu esse proident irure exercitation. Ut do nostrud reprehenderit excepteur nisi eu aute aute qui. Deserunt tempor sint veniam enim Lorem laboris. Commodo id exercitation magna aliquip cillum elit.
Lorem culpa ea ad
Excepteur anim officia anim sit consectetur ex aliqua amet exercitation incididunt eu deserunt cupidatat. In enim pariatur magna non anim occaecat. Et aliquip pariatur dolor nulla eu anim enim incididunt eiusmod minim.
Login to see this section
About this result
This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.
Related lectures (40)
SGD and Mean Field Analysis
Explores Stochastic Gradient Descent and Mean Field Analysis in two-layer neural networks, emphasizing their iterative processes and mathematical foundations.
Provable Benefits of Overparameterization in Model Compression
Explores the provable benefits of overparameterization in model compression, emphasizing the efficiency of deep neural networks and the importance of retraining for improved performance.
Optimization for Machine Learning: Non-convex
Explores non-convex optimization in machine learning, covering gradient descent, trajectory analysis, and linear models.
Generalization in Deep Learning
Explores generalization in deep learning, covering model complexity, implicit bias, and the double descent phenomenon.
Stochastic Gradient Descent
Explores stochastic gradient descent optimization and the Mean-Field Method in neural networks.
Show more