Difference between revisions of "Softmax function"
Jump to navigation
Jump to search
(Created page with "wikipedia:Softmax function") Tags: Mobile web edit, Mobile edit |
Tags: Mobile web edit, Mobile edit |
||
| (3 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
| − | [[wikipedia:Softmax function]] | + | [[wikipedia:Softmax function]] [[softargmax]] |
| + | |||
| + | The softmax function is often used as the last activation function of a neural network to normalize the output of a network to a probability distribution over predicted output classes. | ||
| + | |||
| + | A smooth approximation to the [[maximum]] function | ||
| + | |||
| + | |||
| + | * [[Cross-entropy]] | ||
| + | |||
| + | {{NN}} | ||
Latest revision as of 06:48, 10 September 2025
wikipedia:Softmax function softargmax
The softmax function is often used as the last activation function of a neural network to normalize the output of a network to a probability distribution over predicted output classes.
A smooth approximation to the maximum function
Artificial neural networks, Neuronal network (NN), CNN, RNN, Micrograd, NPU, ConvNet, AlexNet, GoogLeNet, Apache MXNet, Neural architecture search, DAG, Feedforward neural network, NeurIPS, Feature Pyramid Network, TPU, NPU, Apple Neural Engine (ANE), LLM, TFLOPS, Softmax function, Dilution (neural networks), AlphaGo
Advertising: