Learning Mackey Glass from examples Plus or Minus

We apply active exemplar selection Plutowski White to predicting a chaotic time series Given a xed set of ex amples the method chooses a concise subset for training Fitting these exemplars results in the entire set being t as well as de sired The algorithm incorporates a method for regulating network complexity automatically adding exemplars and hidden units as needed Fitting examples generated from the Mackey Glass equa tion with fractal dimension to an rmse of required about exemplars and to hidden units The method requires an order of magnitude fewer oating point operations than training on the entire set of examples is signi cantly cheaper than two contend ing exemplar selection techniques and suggests a simpler active selection technique that performs comparably