Reducing Overfitting Problem in Machine Learning using L1/4 Activation Function

Abstract

Abstract:Machine learning has various applications.
Machine learning model has two problems.Overfittingand
Underfitting.Underfitting is a statistical model or a machine
learning algorithm,it cannot capture the underlying trend of
the data.A statistical model is said to be overfitted, when it is
trained with a lot of data.When model has trained on fewer
features, the machine will be too biased, and then the model
gets underfitting problem. So, it is needed to train the model
on more features and there is one more problem
occurs.Overfitting problem can be reduced by using
regularization functions and data augmentation. In the
previous research on activation functions, Hock Hung
Chieng Proposed an activation function called Flatten-T
Swish: a thresholdReLU, which is a multiplication of Relu
and sigmoid function.
Index Terms: Flatten-T swish, Machine learning, activation
function