Today was the 4th round of reading the book Pattern Recognition and Machine learning by CM Bishop during the student seminar. Inafuku san delivered the presentation this time, discussing various approaches for estimation of probability densities.
We discussed parametric and non-parametric approaches today, taking examples of Gaussian distribution and Exponential distributions for the former and how they influence the predictive power of our model.
Sato san had some valuable inputs by the end of seminar where he described how non-informative priors are not influenced by the data and can be used for more general cases, without influencing the posterior distribution heavily.
Nitish Jaiswal