DigiNews

Tech Watch by Johan Denoyer

← Back to articles

Jensen–Shannon divergence

Quality: 8/10 Relevance: 8/10

Summary

Jensen–Shannon divergence is a symmetric, finite measure of similarity between probability distributions, defined from the KL divergence using the mixture distribution M = (P + Q) / 2. It has a quantum generalization, bounds between 0 and 1 for base-2 logs, and links to mutual information. The article covers definitions, bounds, relation to mutual information, centroid, applications, notes, and external tools.

🚀 Service construit par Johan Denoyer