arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.09191cs.LGstat.ML
收藏

基于加权再训练的深层生成模型潜在空间样本有效优化

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

University of Cambridge(剑桥大学);Alan Turing Institute(艾伦·图灵研究所);Microsoft Research(微软研究院)

Austin Tripp, Erik Daxberger, José Miguel Hernández-Lobato

摘要:

科学和工程中的许多重要问题,如药物设计,涉及在复杂的,高维的,结构化的输入空间上优化昂贵的黑箱目标函数。尽管机器学习技术在解决这些问题方面已经显示出前景,但是现有方法基本上缺乏样本效率。我们介绍了一种改进的高效黑盒优化方法,该方法在通过深度生成模型学习的低维连续隐流形中进行优化。与以前的方法相比,我们积极地引导生成模型来维护一个潜在的流形,这对于有效地优化目标非常有用。我们通过周期性地重新训练生成模型来实现这一点,这些生成模型是沿着优化轨迹查询的数据点,并根据目标函数值对这些数据点进行加权。这种加权再训练可以很容易地在现有方法的基础上实现,并且经验表明,它可以显着提高合成和现实世界优化问题的效率和性能。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑