What is the difference between Perceptron and MLP ?
Welcome, @jesus_david_hurtado. By MLP I assume that you are using an acronym for “multi-layer perceptron.” At an elementary level, you can think of both as binary classifiers. Both are terms that were coined decades ago, at the dawn of AI. Interestingly, some authors continue to refer to multi-layer neural networks as MLPs.
But when the terms were first applied, there was very little in the way of explicit nonlinearities as would be captured by modern activations functions. In these terms, the “activation function” was a Heaviside function: if z>0, then the output equaled one, i.e. the neuron was activated. If z<=0, the output was zero (the neuron was not active).
You can read up on some of the history in the Wikipedia entry on the perceptron.