Latent Dirichlet Allocation (LDA) is a generative model which produces a list of topics. Each topic is represented by a distribution over words.
What is the objective function that it tries to minimize? Can someone explain it in simpler way? The mathematical formula and the intuition behind it?
Secondly, After building the LDA model how we validate the model? I have heard perplexity as one metric, but what is perplexity about?
Does bias, Variance problem also exists in LDA? If yes, when it can occur?