Sharif Digital Repository

In these days world, the increasing growth of Nanotechnology has caused to invent and create new numerical and also computational methods which have more abilities and capabilities for evaluating systems in this scale. Although Some techniques, such as Molecular Dynamics Methods are capable of evaluating nanostructures, lack the ability to simulate large systems of practical size and time scales which is the most important index during the simulation. Therefore, in order to be able to produce an acceptable exact simulation of a large model, simulation of which is limited by the computational cost of the current molecular dynamics methods at hand, Coarse-Graining technique has recently become... 

Nowadays, methods based on deep learning networks are the most effective artificial in­ telligence methods. Although they have achieved success in various fields (such as machine vision and object recognition), practical and experimental cases show the fragility of deep learning networks against perturbations and unwanted changes of the input pattern. All these perturbations must be in a way that the main class of the perturbed input pattern can be rec­ ognized by human, but the network makes a mistake in recognizing its correct class. This thesis seeks a more accurate evaluation by designing adversarial attacks such that the main class of the adversarial pattern is detectable by human...