Generalized activation function for machine learning

The invention relates to a generalized activation function for machine learning. A machine learning model is provided in which each activation node within the model has an adaptive activation function defined by inputs and hyper-parameters of the model. Thus, each activation node may have a separate, different activation function based on the adaptive activation function, where the hyper-parameter of each activation node is trained during the overall training of the model. Furthermore, the disclosure provides that a set of adaptive activation functions can be provided for each activation node, so that an activated spike sequence can be generated. 本公开涉及用于机器学习的广义激活函数。提供了一种机器学习模型，其中该模型内的每个激活节点具有按照模型的输入和超参数定义的自适应激活函数。因此，每个激活节点基于自适应激活函数可具有单独的不同激活函数，其中每个激活节点的超参数在模型的整体训练期间被训练。此外，本公开提供了，可以为每个激活节点提供一组自适应激活函数，从而可以生成激活的尖峰序列。