A New Approach of Deriving Bounds between Entropy and Error from Joint Distribution: Case Study for Binary Classifications
Bao-Gang HuHong-Jie XingPublished in: CoRR (2013)
Keyphrases
- joint distribution
- case study
- error bounds
- maximum entropy
- random variables
- marginal distributions
- minimum error
- lower bound
- marginal probabilities
- conditional distributions
- information theory
- conditional probabilities
- information theoretic
- upper bound
- conditional distribution
- probability distribution
- upper and lower bounds
- error tolerance
- real world
- lower and upper bounds
- expected error
- generalization error
- worst case
- multiscale
- adaptive binary arithmetic coding