Constraints or preferences that guide generalization beyond observed training examples. Local convolution, weight sharing, a compositional grammar and a prior over programs supply different inductive biases. Their usefulness depends on the structure of the target problem.
New to topics? Read the docs here!