This project is about learning how to ground quantified goals in classical planning. The quantified goals can not be handled by planners out-of-the box. That the goals are quantified means that there is at least one existential quantifier (from first order logic) in the goal. This can be seen as the object in the goal only being referred to by its properties, without giving the ID of the object. Therefore, we try to learn the function that goes from a quantified goal to a ground goal, which can then be used as the input to a classical planner for finding the sequence of actions that takes the agent from the initial state to the goal state.