Evaluating the Effects of Convolutional Neural Network Committees (CROSBI ID 633739)
Prilog sa skupa u zborniku | izvorni znanstveni rad | međunarodna recenzija
Podaci o odgovornosti
Jurišić, Fran ; Filković, Ivan ; Kalafatić, Zoran
engleski
Evaluating the Effects of Convolutional Neural Network Committees
Many high performing deep learning models for image classification put their base models in a committee as a final step to gain competitive edge. In this paper we focus on that aspect, analyzing how committee size and makeup of models trained with different preprocessing methods impact final performance. Working with two datasets, representing both rigid and non- rigid object classification in German Traffic Sign Recognition Benchmark (GTSRB) and CIFAR- 10, and two preprocessing methods in addition to original images, we report performance improvements and compare them. Our experiments cover committees trained on just one dataset variation as well as hybrid ones, unreliability of small committees of low error models and performance metrics specific to the way committees are built. We point out some guidelines to predict committee behavior and good approaches to analyze their impact and limitations.
machine learning ; deep learning ; convolutional neural networks ; classification
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
560-565.
2016.
objavljeno
Podaci o matičnoj publikaciji
Vol. 4: VISAPP - International Conference on Computer Vision Theory and Applications
Battiato, Sebastiano ; Imai, Francisco
Setúbal: SCITEPRESS
978-989-758-175-5
Podaci o skupu
11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2016)
predavanje
27.02.2016-29.02.2016
Rim, Italija