Gradient Descent Models Are Kernel Machines (Deep Learning)

Retrieved on: 2021-02-13 16:07:30

Tags for this article:

Academic disciplines

Learning

Computational statistics

Hilbert space

Computational neuroscience

Machine learning

Positive-definite kernel

Kernel method

Artificial neural network

Deep learning

Multiple kernel learning

Kernel embedding of distributions

Click the tags to see associated articles and topics

Gradient Descent Models Are Kernel Machines (Deep Learning). View article details on hiswai:

Excerpt

The kernels are defined by the inner product of model gradients in the parameter space, integrated over the descent (learning) path. Roughly speaking, ...

Article found on: www.predictiveanalyticsworld.com

View Original Article