The method [28] reviewed in this section is an instance of a strategy known as ``predictive coding'' or ``model-based coding''. A neural predictor network is trained to approximate the conditional probability distribution of possible characters, given the previous characters. 's outputs are fed into the Arithmetic Coding algorithm (e.g. [41]) that generates short codes for characters with low information content (characters with high predicted probability) and long codes for characters conveying a lot of information (highly unpredictable characters) [28].

- PREDICTING CONDITIONAL PROBABILITIES
- USING THE PREDICTOR FOR COMPRESSION
- SIMULATIONS
- ON-LINE METHODS / LIMITATIONS

Juergen Schmidhuber 2003-02-19