| Summary: | <p> The inverse problem is one of the classical problems in Computer Science. There are currently several numerical solutions for this problem based on Linear Algebra. Typically, the forward problem is when we know a model, or a formula, and we compute the values. On the contrary, the inverse problem is when the data is collected with some measuring equipment and then inverted to find the model. It can be described as identifying the cause using its effect. However, there may not exist a unique solution to this problem, but there are approximations to guess what the information might have been. These methods suffer from downsides, because there is not enough data to compute an appropriate solution. This thesis presents a possible approach to the inverse problem using Machine Learning for the Electroencephalography (EEG) dataset and presents an analysis of the results obtained by testing some of the known Unsupervised Learning methods.</p><p>
|
|---|
