What are Neural Networks made of?

08/25/2019
by   Rene Schaub, et al.
0

The success of Deep Learning methods is not well understood, though various attempts at explaining it have been made, typically centered on properties of stochastic gradient descent. Even less clear is why certain neural network architectures perform better than others. We provide a potential opening with the hypothesis that neural network training is a form of Genetic Programming.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/22/2022

Langevin algorithms for Markovian Neural Networks and Deep Stochastic control

Stochastic Gradient Descent Langevin Dynamics (SGLD) algorithms, which a...
research
07/03/2019

Neural Network Architecture Search with Differentiable Cartesian Genetic Programming for Regression

The ability to design complex neural network architectures which enable ...
research
02/08/2021

Neurogenetic Programming Framework for Explainable Reinforcement Learning

Automatic programming, the task of generating computer programs complian...
research
07/26/2022

Quiver neural networks

We develop a uniform theoretical approach towards the analysis of variou...
research
05/04/2023

A Bootstrap Algorithm for Fast Supervised Learning

Training a neural network (NN) typically relies on some type of curve-fo...
research
10/23/2019

Autoencoding with XCSF

Autoencoders enable data dimensionality reduction and are a key componen...
research
12/09/2017

Peephole: Predicting Network Performance Before Training

The quest for performant networks has been a significant force that driv...

Please sign up or login with your details

Forgot password? Click here to reset