LaplaceNet: A Hybrid Graph-Energy Neural Network for Deep Semisupervised Classification.
Philip SellarsAngelica I. Avilés-RiveroCarola-Bibiane SchönliebPublished in: IEEE Trans. Neural Networks Learn. Syst. (2024)
Keyphrases
- neural network
- pattern recognition
- learning vector quantization
- semi supervised
- semi supervised classification
- machine learning
- pattern classification
- unsupervised learning
- random walk
- image classification
- classification accuracy
- decision trees
- supervised learning
- classification scheme
- incremental learning
- feature extraction
- energy consumption
- fault diagnosis
- fuzzy logic
- multi layer perceptron
- classification models
- bp neural network
- class labels
- support vector machine svm
- training set
- artificial neural networks
- back propagation
- classification algorithm
- training data
- energy minimization
- text classification
- markov random field
- feature vectors
- graph representation
- active learning
- genetic algorithm
- transductive learning
- kohonen self organizing map