Comparative Performance Analysis of Neural Networks Architectures on H2O Platform for Various Activation Functions

TL;DR

This paper evaluates different neural network architectures and activation functions on the H2O platform using the MNIST dataset, highlighting how parameter choices impact runtime and accuracy in image recognition tasks.

Contribution

It provides empirical insights into how parameter tuning affects neural network performance and efficiency on the H2O platform for image recognition.

Findings

Parameter selection can increase runtime by 2-3 orders of magnitude.

Blind parameter tuning may not significantly improve accuracy.

Optimization strategies are crucial for efficient deep learning applications.

Abstract

Deep learning (deep structured learning, hierarchi- cal learning or deep machine learning) is a branch of machine learning based on a set of algorithms that attempt to model high- level abstractions in data by using multiple processing layers with complex structures or otherwise composed of multiple non-linear transformations. In this paper, we present the results of testing neural networks architectures on H2O platform for various activation functions, stopping metrics, and other parameters of machine learning algorithm. It was demonstrated for the use case of MNIST database of handwritten digits in single-threaded mode that blind selection of these parameters can hugely increase (by 2-3 orders) the runtime without the significant increase of precision. This result can have crucial influence for opitmization of available and new machine learning methods, especially for image recognition problems.