Machine Learning and Generative adversarial network enter art scene in the last decade. As we can see, there are many different applications for this artificial intelligence approach to learn new tasks: search algorithms, financial prediction systems etc... A basic algorithm is written from scratch by a trained neural network - sometimes with parameters already defined (e.g., learning rate) or learned using deep reinforcement learning techniques such as backpropagation training where an agent learns via its own actions when it receives information about how hard was challenging enough at any given time. This type of algorithmic model requires only inputs like data stream size, dimensions, accuracy threshold, number that needs processing during execution and so on but not much more since.