Scalable Pareto Front Approximation for Deep Multi-Objective Learning

Understand

Multi-objective optimization (MOO) is a prevalent challenge for Deep Learning, however, there exists no scalable MOO solution for truly deep neural networks.

• Prior work either demand optimizing a new network for every point on the Pareto front, or induce a large overhead to the number of trainable parameters by using hyper-networks conditioned on modifiable preferences.
• In this paper, we propose to condition the network directly on these preferences by augmenting them to the feature space.
• Furthermore, we ensure a well-spread Pareto front by penalizing the solutions to maintain a small angle to the preference vector.