Table 2 Machine-learning backbones optimized with DF–GDA for every dataset evaluated in the paper

From: A dynamic fractional generalized deterministic annealing for rapid convergence in deep learning optimization

Backbone ML Model

Dataset(s)

LeNet-5 CNN

MNIST

RBF-SVM

MNIST (separate SVM study)

3-layer CNN

MNIST-M, CIFAR-10, SVHN, USPS

ResNet-50 (2-D CNN)

ImageNet

3D-ResNet-50 (Spatiotemp CNN)

Kinetics-700

LSTM RNN

HAR

1-D CNN (text)

IMDB Sentiment, SMS Spam, Airline Sent

Feed-forward NN (+ dropout)

Breast Cancer, Heart Disease, Liver Patient

Feed-forward NN

IRIS, YEAST

Back to article page