Verified80% confidenceFactExact time
KL divergence is asymmetric, i.e., D(p,q) ≠ D(q,p).
4
Sources
80%
Confidence
Long-term
Relevance
8/2/2026
First Seen
Sources
From Entropy to Cross-Entropy: An Information-Theoretic Derivation of the Logistic Regression Loss Function
Related Entities
Related Claims
UnverifiedKL散度是非对称的，P相对于Q的KL散度与Q相对于P的KL散度不相等83% similarUnverifiedKL divergence is asymmetric, D_KL(p||q) ≠ D_KL(q||p), and non-negative, equaling zero if and only if the two distributions are identical.80% similarUnverifiedKL散度最早由Solomon Kullback和Richard Leibler在1951年提出，具有非对称性即KLD(P||Q) ≠ KLD(Q||P)72% similarUnverifiedUnlike KL divergence, the Wasserstein distance has geometry-aware properties and does not diverge to infinity as KL divergence does when the supports do not completely overlap.72% similarUnverifiedJS divergence is a symmetrized version of KL divergence, with results always bounded in [0, ln2], avoiding KL divergence's asymmetry and the defect of diverging to infinity when support sets do not overlap.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/679896API
curl https://kongchang.com/api/v1/knowledge/claims/679896MCP
get_claim(id=679896)