Data samples and estimations on probability distributions

Data in a machine learning project typically comes from a process whose details are not fully known. This lack of knowledge is embraced by modeling the process as a random process. Even if the data is generated through a fully deterministic process, because we do not have a complete knowledge of it, we model it as random and use probability theory to analyse it. This video is part of a full course on Foundations of Machine Learning that is freely available at https://youtu.be/aTRKf3pmv6o. Coding assignments: https://github.com/ionpetre/FoundML_course_assignments