AI engineering glossary · Deep learning core
What is NLL (negative log-likelihood)?
Negative log-likelihood (also called NLL): A loss equal to minus the log of the probability a model gives to the observed data. Minimising it is the same as maximising likelihood, and for classifiers it coincides with cross-entropy when labels are one-hot.
Negative log-likelihood: an example
Three independent predictions with probabilities 0.5, 0.8 and 0.9 give NLL = -(ln 0.5 + ln 0.8 + ln 0.9), about 1.02.
Related terms
- Cross-entropy loss
- Loss function
- Softmax
More Deep learning core terms
- Neural network pruning
- Neural tangent kernel
- Neuron
- One-cycle policy
- Optimiser
- Orthogonal initialisation
- Output layer
- Parametric ReLU
Where you learn it
Negative log-likelihood is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.