Title :
Bayesian geometric theory of learning algorithms
Author_Institution :
Santa Fe Inst., NM, USA
Abstract :
The problem of objective evaluation of learning algorithms is analyzed under the principles of coherence and covariance. The theory of Bayesian information geometry satisfies these principles and encompasses most of the commonly used learning criteria. Implications to learning theory are discussed
Keywords :
Bayes methods; computational geometry; estimation theory; information theory; learning (artificial intelligence); optimisation; probability; Bayesian geometric theory; Bayesian information geometry; coherence; covariance; learning algorithms; optimal estimators; probability; Algorithm design and analysis; Bayesian methods; Equations; Information geometry; Iron; Least squares methods; Neural networks; Roads; Statistics; Transforms;
Conference_Titel :
Neural Networks,1997., International Conference on
Conference_Location :
Houston, TX
Print_ISBN :
0-7803-4122-8
DOI :
10.1109/ICNN.1997.616171
