—Probability models are estimated by use of penalized log-likelihood criteria related to AIC and MDL. The accuracies of the density estimators are shown to be related to the trad...
In recent years, gradient-based LSTM recurrent neural networks (RNNs) solved many previously RNN-unlearnable tasks. Sometimes, however, gradient information is of little use for t...
A coevolutionary competitive learning environment for two antagonistic agents is presented. The agents are controlled by a new kind of computational network based on a compartment...
Gul Muhammad Khan, Julian Francis Miller, David M....
Communication overheads are one of the fundamental challenges in a multiprocessor system. As the number of processors on a chip increases, communication overheads and the distribu...
Katherine E. Coons, Behnam Robatmili, Matthew E. T...
A neuro-fuzzy methodology is described which involves connectionist minimization of a fuzzy feature evaluation index with unsupervised training. The concept of a ¯exible membersh...