Triplet Loss

#deep-learning #loss-in-ml #interview

Triplet loss is the improvement over the Contrastive Loss
Triplet loss expects one anchor, one positive and one negative point
It simultaneously minimizes the Euclidian Distance between positive and anchor
AND maximizes the Euclidian Distance between negative and anchor
As it works simultaneously it needs less amount of sample for convergence

Triplet Loss Formula

$l o s s = m a x (0, | x - x^{+} |^{2} - | x - x^{-} |^{2} + m)$

Related Notes