Gaussian-Based Runtime Detection of Out-of-distribution Inputs for Neural Networks.
Vahid HashemiJan KretínskýStefanie MohrEmmanouil SeferisPublished in: RV (2021)
Keyphrases
- neural network
- gaussian distribution
- poisson distribution
- input pattern
- detection accuracy
- artificial neural networks
- normal distribution
- detection algorithm
- generalized likelihood ratio
- marginal distributions
- gaussian model
- false positives
- detection method
- object detection
- probability distribution
- pattern recognition
- independent and identically distributed
- self organizing maps
- mixture of gaussians
- gaussian mixture model
- detection rate
- neyman pearson
- power law
- density function
- false alarms
- probability density
- neural network model
- joint distribution
- fuzzy systems
- neural nets
- feed forward
- automatic detection
- event detection
- data distribution
- random variables
- back propagation
- fuzzy logic
- control system
- learning algorithm