IMPROVISASI BACKPROPAGATION MENGGUNAKAN PENERAPAN ADAPTIVE LEARNING RATE DAN PARALLEL TRAINING
DOI:
https://doi.org/10.29103/techsi.v6i1.169Abstract
Artificial neural networks have long been used in the classification process, which offers the flexibility of neural networks to the features of the object to be classified and small storage space. The biggest drawback of the backpropagation network is the time taken by the network to learn to be very long for large data conditions of learning and the conditions in which the features between different objects have small differences. To overcome the weaknesses of the implementation of the development is carried out by applying the concept of parallel adaptvie learning rate and training in order to improve the ability of the network in the learning process.References
Aan Tri Wibowo. (2013). Pembuatan Aplikasi E-Commerce Pusat
Oleh-Oleh Khas Pacitan Pada Toko Sari Rasa Pacitan. IJNS -
Indonesian Journal on Networking and Security - ISSN: 23025700.
Moreira, M., & Fiesler, E. (1995). Neural Network with Adaptive Learning Rate and Momentum Terms. SUISSE: Institut Dalle Molle D'Intelligence Artificelle Perceptive.
Mumtazimah, M. (2012). Parallel Training for Back Propagation in Character Recognition. Terengganu: ICCIT.
Plagianakos, V. P. (1998). An Improved Backpropagation Method with Adaptive Learning Rate. Patras: University Of Patras.
Putra, D. (2010). Pengolahan Citra Digital. Yogyakarta: Penerbit Andi.
Schuessler, O., & Loyola, D. (2011). Parallel Training of Artificial Neural Network Using Multithreaded and Multicore CPUs. Berlin: German Aerospace Center, Institute of Remote Sensing.
Downloads
Published
Issue
Section
License
Authors retain copyright and grant the journal right of first publication and this work is licensed under a Creative Commons Attribution-ShareAlike 4.0 that allows others to share the work with an acknowledgement of the works authorship and initial publication in this journal.
All articles in this journal may be disseminated by listing valid sources and the title of the article should not be omitted. The content of the article is liable to the author.
Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
In the dissemination of articles by the author must declare the TECHSI Journal as the first party to publish the article.
