Online Program Home
My Program

Abstract Details

Activity Number: 253 - Contributed Poster Presentations: Quantum Computing in Statistics and Machine Learning
Type: Contributed
Date/Time: Monday, July 29, 2019 : 2:00 PM to 3:50 PM
Sponsor: Quantum Computing in Statistics and Machine Learning
Abstract #306421
Title: Optimization of Backpropagation Multilayer Neural Network
Author(s): Jun Kim* and Anindya Bhadra
Companies: Purdue University and Purdue University
Keywords: Machine Learning; Neural Network
Abstract:

While there has been a growing interest of optimization implements into a neural network, one known method are mini-batches. While batch gradient descent may result in a more stable convergence on some problems, it can also result in a premature convergence of a model. A mini-batch gradient descent allows a more robust convergence, which may allow the neural network to avoid local minima. This poster will explore the optimum number of mini-batch sizes on different sizes of data set by using a gradient descent multilayer backpropagation model.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program