Sample-Efficient Optimization in the Latent Space of Deep Generative Models via Weighted Retraining

By Austin Tripp, Erik A. Daxberger, José Miguel Hernández-Lobato. 2020

175 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

In Computer Vision and Pattern Recognition

Citations in the corpus

Getting a CLUE: A Method for Explaining Uncertainty Estimates (2020, background): “An alternative more amenable to high dimensional data is to leverage DGMs (which we dub auxiliary DGMs ) to ensure explanations are in-distribution (Dhurandhar et al., 2018; Joshi et al., 2018; Chang et al., 2019; Booth et al., 2020; Tripp et al., 2020).”
Guided Generative Protein Design using Regularized Transformers (2022, background): “In contrast to Tripp et al. [2020] and Brookes et al. [2019], our model only requires training once and learns a latent space where high-ﬁtness samples gradated away from low-ﬁtness samples.”