Forward modeling is an important step in processing data of seabed logging (SBL) with controlled source electromagnetic (CSEM) method to determine the location of a hydrocarbon layer under the seafloor. In this research, forward modeling was conducted using a multi layer perceptron (MLP), which is an important type of artificial neural networks. To train this MLP, a data set was generated using simulation software: COMSOL Multiphysics. The MLP designed has 3 layers with 3 neurons in input layer and 1 neuron in output layer. The single hidden layer contained neurons whose number had been varied between 3 until 15 neurons. The performance comparison showed that the MLP with 10 neurons in its hidden layer was the best to model SBL with CSEM method. © 2010 IEEE.