Most existing recommender systems are limited to single type behaviour modelling whereas in real-world scenarios, user behaviour is  multi-type, including activities such as  view, add-to-cart and purchase. Moreover, most state-of-the-art methods fall short in modelling higher-order user-item interactions and long-range dependencies. The focus of this project is to develop efficient  Hypergraph Representation Learning methods with a goal to address the highlighted challenges at lower computational cost.