Researcher, CSIRO
1 paper at NeurIPS 2025
We learn a generative model of the Pareto set that can be conditioned on subjective preferences, without retraining, for online multi-objective optimization tasks on discrete/mixed spaces.