Boltzmann Tuning of Generative Models
This addresses the need for efficient post-hoc tuning of generative models in applications like conditional modeling and robust design, though it appears incremental as an optimization-based extension.
The paper tackles the problem of tuning generative models to favor instances that meet an external differentiable criterion, proposing Boltzmann Tuning of Generative Models (BTGM) as an affordable alternative to rejection sampling. It demonstrates BTGM's ability to sample extreme regions in a real-world energy policy application.
The paper focuses on the a posteriori tuning of a generative model in order to favor the generation of good instances in the sense of some external differentiable criterion. The proposed approach, called Boltzmann Tuning of Generative Models (BTGM), applies to a wide range of applications. It covers conditional generative modelling as a particular case, and offers an affordable alternative to rejection sampling. The contribution of the paper is twofold. Firstly, the objective is formalized and tackled as a well-posed optimization problem; a practical methodology is proposed to choose among the candidate criteria representing the same goal, the one best suited to efficiently learn a tuned generative model. Secondly, the merits of the approach are demonstrated on a real-world application, in the context of robust design for energy policies, showing the ability of BTGM to sample the extreme regions of the considered criteria.