What is label smoothing and why does it help?

easy

Answer

$Replace one-hot target [0, ..., 1, ..., 0] with (1-α)*one-hot + α/K (uniform), typically α=0.1.$
$Prevents the network from becoming over-confident: it can't drive the softmax logit for the true class to infinity because the target isn't exactly 1.$
$Improves calibration, reduces overconfidence, and often improves accuracy 0.5-1% on classification benchmarks.$
$Standard in ImageNet training and NMT.$

Check yourself — multiple choice

Same as dropout
Soften one-hot targets by mixing with uniform — reduces overconfidence, improves calibration and accuracy
Adds noise to features
Only for regression

Label smoothing: (1-α)·onehot + α/K uniform → less overconfident predictions.

#regularization#losses

214 interview questions in this topic.