Description: Machine learning (ML) is a branch of computer science that deals with solving problems by learning through experience. In the classical setup, a human defines all the steps necessary for a computer to solve the problem. However, for complex tasks when it is not trivial to come up with a model to map the input data to our desired output, it is often desirable to learn from the data itself. This process of learning through experience is called "training" an algorithm.
In this tutorial, I will cover topics like what a neural network is, how does it learn, most commonly used architectures (for e.g. Multilayer perceptron, Convolutional neural network, Generative adversarial network) and together we will build a few simple models in order to classify between pulsar signals and RFI/noise. If time permits, we will together try to tackle a mysterious and more light-hearted problem using ML.
Participants: Ferdinand, Arunima, Jompoj, Jaswanth, Tasha, Joscha, Miquel, Jiwoong and Mary
