The IB problem on (T,C) is exactly equivalent to the IB problem on (φ(T),C) whenever φ is a sufficient statistic for C given T, with the full curve and optimal representations preserved.
Information- based clustering.Proceedings of the National Academy of Sciences of the United States of America, 102(51):18297–18302, 2005
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IT 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Sufficient-Statistic Reduction of the Information Bottleneck to a Low-Dimensional Problem
The IB problem on (T,C) is exactly equivalent to the IB problem on (φ(T),C) whenever φ is a sufficient statistic for C given T, with the full curve and optimal representations preserved.