Back to Search
Start Over
The Compact Support Neural Network
- Source :
- Sensors, Vol 21, Iss 8494, p 8494 (2021), Sensors; Volume 21; Issue 24; Pages: 8494, Sensors (Basel, Switzerland)
- Publication Year :
- 2021
- Publisher :
- MDPI AG, 2021.
-
Abstract
- Neural networks are popular and useful in many fields, but they have the problem of giving high confidence responses for examples that are away from the training data. This makes the neural networks very confident in their prediction while making gross mistakes, thus limiting their reliability for safety-critical applications such as autonomous driving, space exploration, etc. This paper introduces a novel neuron generalization that has the standard dot-product-based neuron and the {\color{black} radial basis function (RBF)} neuron as two extreme cases of a shape parameter. Using a rectified linear unit (ReLU) as the activation function results in a novel neuron that has compact support, which means its output is zero outside a bounded domain. To address the difficulties in training the proposed neural network, it introduces a novel training method that takes a pretrained standard neural network that is fine-tuned while gradually increasing the shape parameter to the desired value. The theoretical findings of the paper are a bound on the gradient of the proposed neuron and a proof that a neural network with such neurons has the universal approximation property. This means that the network can approximate any continuous and integrable function with an arbitrary degree of accuracy. The experimental findings on standard benchmark datasets show that the proposed approach has smaller test errors than state-of-the-art competing methods and outperforms the competing methods in detecting out-of-distribution samples on two out of three datasets.<br />Comment: 13 pages, 6 figures
- Subjects :
- Neurons
FOS: Computer and information sciences
Computer Science - Machine Learning
Quantitative Biology::Neurons and Cognition
RBF networks
Chemical technology
universal approximation
Reproducibility of Results
TP1-1185
neural networks
Biochemistry
Article
Atomic and Molecular Physics, and Optics
Machine Learning (cs.LG)
Analytical Chemistry
Benchmarking
OOD detection
Neural Networks, Computer
Electrical and Electronic Engineering
Instrumentation
Algorithms
Subjects
Details
- Language :
- English
- ISSN :
- 14248220
- Volume :
- 21
- Issue :
- 8494
- Database :
- OpenAIRE
- Journal :
- Sensors
- Accession number :
- edsair.doi.dedup.....d6f85d8dbd63696737482d6c06dc13cb