These days studies in astronomy might require numerous calculations and simulations. While one can resolve many issues by using classical numerical and analytical methods, this might not be efficient for problems involving big data. Moreover, some tasks might be time consuming in terms of Human Resources what make them cost-inefficient and hence complicated.
By using machine-learning methods, one can improve the overall performance of studies requiring either high volume computations or human agents to be performed. In this speech I make a brief overview of some existing applications of ML methods in astronomy, review in details some research performed in this area, provide a classification of tasks one might resolve by using ML, and review the existing tools, which one can easily start using without prior background.
Below is the keynote of my talk at the Astronomical observatory of Belgrade.