Anthropic Correction of Information Estimates

A novel estimator for mutual information is proposed. The estimator is useful for the (asymmetric) scenario where only a few samples for one random variable are available, but for each sample, the conditional distribution of the other random variable can be accurately characterized. Such asymmetry is common in neuroscience where it is often necessary to repeat the same stimulus many times to obtain a stable response.


Published in:
Itw: 2009 Ieee Information Theory Workshop On Networking And Information Theory, 152-155
Presented at:
IEEE Information Theory Workshop on Networking and Information Theory, Volos, GREECE, Jun 10-12, 2009
Year:
2009
Publisher:
Ieee Service Center, 445 Hoes Lane, Po Box 1331, Piscataway, Nj 08855-1331 Usa
ISBN:
978-1-4244-4535-6
Laboratories:




 Record created 2011-10-17, last modified 2018-03-17


Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)