We propose a novelapproach to automaticallygrowing and pruning Hierarchical Mixtures of Experts. The constructive algorithm proposed here enables large hierarchies consisting of s...
Real-valued random hidden variables can be useful for modelling latent structure that explains correlations among observed variables. I propose a simple unit that adds zero-mean G...
Dynamic Programming, Q-learning and other discrete Markov Decision Process solvers can be applied to continuous d-dimensional state-spaces by quantizing the state space into an arr...
Predictions oflifetimesofdynamicallyallocated objects can be used to improve time and space e ciency of dynamic memory management in computer programs. Barrett and Zorn 1993] used...
Support Vector Learning Machines (SVM) are nding application in pattern recognition, regression estimation, and operator inversion for ill-posed problems. Against this very genera...