Efficient Methods for Dealing with Missing Data in Supervised Learning (2002)
In G. Tesauro, D. S. Touretzky, T. K. Leen, Advances In, Volker Tresp, Ralph Neuneier, ...
We present efficient algorithms for dealing with the problem of missing inputs (incomplete feature vectors) during training and recall.
Combining Estimators Using Non-Constant Weighting Functions (2002)
In G. Tesauro, D. S. Touretzky, T. K. Leen, Advances In, Volker Tresp
This paper discusses the linearly weighted combination of estimators in which the weighting functions are dependent on the input. We show that the weighting functions can be derived...
Tree-Based Modeling and Estimation of Gaussian Processes on Graphs with Cycles (2001)
Martin J. Wainwright, Erik B. Sudderth, Alan S. Willsky, T. K. Leen, T. G. Dietterich, V. Tresp
We present the embedded trees algorithm, an iterative technique for estimation of Gaussian processes defined on arbitrary graphs. By exactly solving a series of modified problems on embedded spanning...
New Approaches Towards Robust and Adaptive Speech Recognition (2001)
Bourlard, H, Bengio, S., Weber, K., Leen, T. K., Dietterich, T. G., Tresp, V.
Understanding stepwise generalization of Support Vector Machines: a toy model (2000)
S. A. Solla, T. K. Leen, Sebastian Risau-gusman, Mirta B. Gordon
In this article we study the eects of introducing structure in the input distribution of the data to be learnt by a simple perceptron. We determine the learning curves within the framework of...
Learning Local Error Bars for Nonlinear Regression (1997)
D. S. Touretzky, T. K. Leen, David A. Nix, Andreas S. Weigend
We present a new method for obtaining local error bars for nonlinear regression, i.e., estimates of the confidence in predicted values that depend on the input. We approach this problem by applying a...
Resonance in a Stochastic Neuron Model with Delayed Interaction (1970)
S. A. Solla, T. K. Leen, K. R. Muller, Toru Ohira, Yuzuru Sato, Jack D. Cowan
We study here a simple stochastic single neuron model with delayed self--feedback capable of generating spike trains. Simulations show that its spike trains exhibit resonant behavior between "noise"...