Loss function
A loss function is a mathematical function commonly used in statistics. Loss functions are frequently used to create machine learning algorithms.
Loss functions are used in machine learning to measure the mathematical distance between predicted values and actual real values.
For example in image classification if an image of the color red is classified as an image of the color red then the loss function is 0; however, if it is classified as an image of blue then the loss function is greater than zero. Any correctly chosen loss function for this example will give a greater value for blue than orange or pink.
The optimization of backpropagation in artificial neural nets is dependent upon loss functions.
There are many types of loss functions including mean absolute loss, mean squared error and mean bias error.
Related Radiopaedia articles
Artificial intelligence
 artificial intelligence (AI)
 imaging data sets
 computeraided diagnosis (CAD)
 natural language processing
 machine learning (overview)
 visualizing and understanding neural networks
 common data preparation/preprocessing steps
 DICOM to bitmap conversion
 dimensionality reduction
 scaling
 centering
 normalization
 principal component analysis
 training, testing and validation datasets
 augmentation
 loss function

optimization algorithms
 ADAM
 momentum (Nesterov)
 stochastic gradient descent
 minibatch gradient descent

regularisation
 linear and quadratic
 batch normalization
 ensembling
 rulebased expert systems
 glossary
 activation function
 anomaly detection
 automation bias
 backpropagation
 batch size
 computer vision
 concept drift
 cost function
 confusion matrix
 convolution
 cross validation
 curse of dimensionality
 dice similarity coefficient
 dimensionality reduction
 epoch
 feature extraction
 gradient descent
 image registration
 imputation
 iteration
 jaccard index
 linear algebra
 noise reduction
 R (Programming language)
 Python (Programming language)
 segmentation
 semisupervised learning
 synthetic and augmented data
 overfitting
 transfer learning