is a personalized information filtering framework designed to identify and present a bundle of items that best match an individual user’s preferences. During Master's course, I conducted research on MLP based latent factor models.

is a deep learning framework designed to model epistemic uncertainty associated with a single reference. During Master's course, I explored how Bayesian deep learning can be applied to handle preference uncertainty in implicit feedback data @ recommendation. 

aims to estimate the probability that a user clicks on an advertisement by modeling the interactions among the user, the advertisement, and the surrounding context.