Activation of bipolar prototypes in attribute inferences

Unsupervised learning of prototypes and attribute weights

In this paper, we introduce new algorithms that perform clustering and feature weighting simultaneously and in an unsupervised manner. The proposed algorithms are computationally and implementationally simple, and learn a di1erent set of feature weights for each identi2ed cluster. The cluster dependent feature weights o1er two advantages. First, they guide the clustering process to partition th...

Unsupervised Learning of Prototypes and Attribute Weights Summary

In this paper, we introduce new algorithms that perform clustering and feature weighting simultaneously and in an unsupervised manner. The proposed algorithms are computationally and implementation ally simple, and learn a different set of feature weights for each identified cluster. The cluster dependent feature weights offer two advantages. First, they guide the clustering process to partitio...

The automatic and co-occurring activation of multiple social inferences

Shifting Mean Activation towards Zero with Bipolar Activation Functions

We propose a simple extension to the ReLU-family of activation functions that allows them to shift the mean activation across a layer towards zero. Combined with proper weight initialization, this alleviates the need for normalization layers. We explore the training of deep vanilla recurrent neural networks (RNNs) with up to 144 layers, and show that bipolar activation functions help learning i...

Shifting Mean Activation Towards Zero with Bipolar Activation Functions

We propose a simple extension to the ReLU-family of activation functions that allows them to shift the mean activation across a layer towards zero. Combined with proper weight initialization, this alleviates the need for normalization layers. We explore the training of deep vanilla recurrent neural networks (RNNs) with up to 144 layers, and show that bipolar activation functions help learning i...