In this talk we provide a general overview of the growing topic of Bayesian nonparametric. The start of BNP can be dated back to 1973 with the definition of the DP from Ferguson, which still remains the first choice for modelling due to its nice conjugacy properties. Since then, the interest has been in extending in several directions the DP in order to overcome its simple structure. We provide a summary of the big picture of BNP and the relationships between the most popular existing priors.
Details about the programme of seminars organised by SMSAS postgraduate research students for SMSAS postgraduate research students.