3 papers
cs.LG2024
DropKAN: Regularizing KANs by masking post-activations
Mohammed Ghaith Altarabichi
We propose DropKAN (Dropout Kolmogorov-Arnold Networks) a regularization method that prevents co-adaptation of activation function weights in Kolmogorov-Arnold Networks (KANs). Dro…
cs.LG2024
Rethinking the Function of Neurons in KANs
Mohammed Ghaith Altarabichi
The neurons of Kolmogorov-Arnold Networks (KANs) perform a simple summation motivated by the Kolmogorov-Arnold representation theorem, which asserts that sum is the only fundamenta…
cs.LG2024
Rolling the dice for better deep learning performance: A study of randomness techniques in deep neural networks
Mohammed Ghaith Altarabichi, SÅawomir Nowaczyk, Sepideh Pashami +2
This paper investigates how various randomization techniques impact Deep Neural Networks (DNNs). Randomization, like weight noise and dropout, aids in reducing overfitting and enha…