AI engineering glossary · Deep learning core
What is BCE (binary cross-entropy)?
Binary cross-entropy (also called BCE): The cross-entropy loss for two-outcome problems: -[y*ln(p) + (1-y)*ln(1-p)], where y is 0 or 1 and p is the predicted probability of class 1. It is normally paired with a sigmoid output.
Binary cross-entropy: an example
For a true label 1 and prediction 0.8 the loss is -ln 0.8, about 0.22; a prediction of 0.2 costs about 1.61.
Related terms
More Deep learning core terms
- Bottleneck
- Catastrophic forgetting
- Cell state
- Checkpoint
- Cross-entropy loss
- Curriculum learning
- Cyclical learning rate
- Data parallelism
Where you learn it
Binary cross-entropy is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.