Multi-objective optimization (MOO) problems are prevalent in machine learning. These problems have a set of optimal solutions, called the Pareto front, where each point on the front represents a different trade-off between possibly conflicting objectives. Most MOO approaches face two grave limitations: (i) A separate model has to be trained for each point on the front; and (ii) The exact trade-off must be known prior to the optimization process. In this work, we propose an approach for learning the entire Pareto front using a single hypernetwork, which we term Pareto HyperNetworks (PHNs). PHN is runtime and memory-efficient compared to training multiple models, and enables the user to select the desired operating point at inference time. * Aviv Navon is a PhD student at Bar Ilan University under the supervision of Prof. Ethan Fetaya and Prof. Gal Chechik. Zoom link: https://technion.zoom.us/j/98665635517