Shannon entropy estimation based on high-rate quantization theory
2015 (English)In: European Signal Processing Conference, European Signal Processing Conference, EUSIPCO , 2015, 1753-1756 p.Conference paper (Refereed)
In this paper we present a method for estimation of the Shannon differential entropy that accounts for embedded manifolds and is based on high-rate quantization theory. It forms an extension of the classical nearest-neighbor entropy estimator and provides simultaneously an explicit estimate of the manifold dimension. A clear advantage of our method over existing methods is that it estimates the more meaningful Shannon entropy rather than the Renyi entropy. Through experiments we confirm the power and the usefulness of our proposed scheme.
Place, publisher, year, edition, pages
European Signal Processing Conference, EUSIPCO , 2015. 1753-1756 p.
Electrical engineering, Differential entropy, Entropy estimator, High-rate quantization, Nearest neighbors, Renyi entropy, Shannon, Shannon entropy, Signal processing
IdentifiersURN: urn:nbn:se:kth:diva-194630ScopusID: 2-s2.0-84979937200OAI: oai:DiVA.org:kth-194630DiVA: diva2:1050421
12th European Signal Processing Conference, EUSIPCO 2004, 6 September 2004 through 10 September 2004
QC 201611292016-11-292016-10-312016-11-29Bibliographically approved