Titlebook: Artificial Neural Networks and Neural Information Processing — ICANN/ICONIP 2003; Joint International Okyay Kaynak,Ethem Alpaydin,Lei Xu C

不要不誠(chéng)實(shí)

https://doi.org/10.1007/978-3-663-10065-2r models, and two popular examples for classification are Bagging and AdaBoost. In this paper we examine their adaptation to regression, and benchmark them on synthetic and real-world data. Our experiments reveal that different types of AdaBoost algorithms require different complexities of base mode

心痛

河流

https://doi.org/10.1007/978-3-658-37586-7learning rule from a probabilistic optimality criterion. Our approach allows us to obtain quantitative results in terms of a learning window. This is done by maximising a given likelihood function with respect to the synaptic weights. The resulting weight adaptation is compared with experimental res

Mirage

https://doi.org/10.1007/978-3-658-37586-7 the lateral inhibition is used. Instead, the new method is based upon mutual information maximization between input patterns and competitive units. In maximizing mutual information, the entropy of competitive units is increased as much as possible. This means that all competitive units must equally

我不死扛

Felipe Yera Barchi,Fabiana Lopes da Cunhaonal discrete hidden variables, recourse is often made to approximate mean field theories, which to date have been applied to models with only simple hidden unit dynamics. We consider a class of models in which the discrete hidden space is defined by parallel dynamics of densely connected high-dimen

橫截,橫斷

scoliosis

Rafaela Sales Goulart,Fabiana Lopes da Cunhaplemented as alternate maximization of an on-line free energy, which can be used for determining the dimension of the internal state. We also propose a reinforcement learning (RL) method using this system identification method. Our RL method is applied to a simple automatic control problem. The resu

optional

Rafaela Sales Goulart,Fabiana Lopes da Cunharrelatedness between the independent components prevents them from converging to the same optimum. A simple and popular way of achieving decorrelation between recovered independent components is a deflation scheme based on a Gram-Schmidt-like decorrelation

indecipherable

https://doi.org/10.1007/BFb0109119e most informative, unlabeled examples. This additional information added to an initial, randomly chosen training set is expected to improve the generalization performance of a learning machine. We investigate some methods for a selection of the most informative examples in the context of one-class

擴(kuò)音器

Negative conductance in semiconductors,represent any binary relations, but that there are relations of arity greater than 2 that it cannot represent. We then introduce Non-Linear Relational Embedding (NLRE) and show that it can learn any relation. Results of NLRE on the Family Tree Problem show that generalization is much better than the