Output Functions

Edit

General • 10 methods

Output functions are layers used towards the end of a network to transform to the desired form for a loss function. For example, the softmax relies on logits to construct a conditional probability. Below you can find a continuously updating list of output functions.

Methods

Add a Method

Method	Year	Papers
Softmax	2000	25318
Heatmap Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation	2014	362
Mixture of Logistic Distributions	2000	158
Adaptive Softmax Efficient softmax approximation for GPUs	2016	66
EVM The Extreme Value Machine	2015	26
Sparsemax From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classification	2016	24
Hierarchical Softmax	2000	9
PAFs Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields	2016	6
Mixture of Softmaxes Breaking the Softmax Bottleneck: A High-Rank RNN Language Model	2017	3

Output Functions Edit

Methods Add a Method

Output Functions

Edit

Methods

Add a Method