Robust Multilayer Perceptrons: Robust Loss Functions and Their Derivatives
Autor: | Petra Vidnerová, Jan Kalina |
---|---|
Rok vydání: | 2020 |
Předmět: |
Artificial neural network
Computer science Computer Science::Neural and Evolutionary Computation Least trimmed squares 02 engineering and technology Perceptron Backpropagation Robust regression 03 medical and health sciences 0302 clinical medicine Multilayer perceptron 0202 electrical engineering electronic engineering information engineering Feedforward neural network 020201 artificial intelligence & image processing Types of artificial neural networks Algorithm 030217 neurology & neurosurgery |
Zdroj: | Proceedings of the 21st EANN (Engineering Applications of Neural Networks) 2020 Conference ISBN: 9783030487904 EANN |
Popis: | Common types of artificial neural networks have been well known to suffer from the presence of outlying measurements (outliers) in the data. However, there are only a few available robust alternatives for training common form of neural networks. In this work, we investigate robust fitting of multilayer perceptrons, i.e. alternative approaches to the most common type of feedforward neural networks. Particularly, we consider robust neural networks based on the robust loss function of the least trimmed squares, for which we express formulas for derivatives of the loss functions. Some formulas, which are however incorrect, have been already available. Further, we consider a very recently proposed multilayer perceptron based on the loss function of the least weighted squares, which appears a promising highly robust approach. We also derive the derivatives of the loss functions, which are to the best of our knowledge a novel contribution of this paper. The derivatives may find applications in implementations of the robust neural networks, if a (gradient-based) backpropagation algorithm is used. |
Databáze: | OpenAIRE |
Externí odkaz: |