Sample-Efficient Optimization in the Latent Space of Deep Generative Models via Weighted Retraining
This addresses the challenge of expensive black-box optimization for applications like drug design, though it is incremental as it builds on existing methods with a novel weighting technique.
The paper tackles the problem of sample-efficient optimization in high-dimensional structured spaces, such as drug design, by introducing a method that actively steers a deep generative model's latent manifold through weighted retraining, resulting in significant improvements in efficiency and performance on synthetic and real-world problems.
Many important problems in science and engineering, such as drug design, involve optimizing an expensive black-box objective function over a complex, high-dimensional, and structured input space. Although machine learning techniques have shown promise in solving such problems, existing approaches substantially lack sample efficiency. We introduce an improved method for efficient black-box optimization, which performs the optimization in the low-dimensional, continuous latent manifold learned by a deep generative model. In contrast to previous approaches, we actively steer the generative model to maintain a latent manifold that is highly useful for efficiently optimizing the objective. We achieve this by periodically retraining the generative model on the data points queried along the optimization trajectory, as well as weighting those data points according to their objective function value. This weighted retraining can be easily implemented on top of existing methods, and is empirically shown to significantly improve their efficiency and performance on synthetic and real-world optimization problems.