Machine learning is becoming an integral part of particle physics. Machine Learning (ML) techniques are emerging as a competitive tool for analyzing and extracting information from large volumes of complex high-dimensional data. In the last few years, the High Energy Physics community has adopted and customized a variety of ML techniques for various steps of data analysis, e.g., event reconstruction, particle identification, jet tagging, signal/background classification, etc. In this talk, I will give a brief introduction to the recent advances in Higgs physics using ML. It includes boosted Higgs tagging, distinguishing Higgs production and decay modes, and self-coupling measurements etc. Thereafter, I will focus on my latest work where we have studied the Higgs tagging performance using ML and theory inspired jet representations and observables.