arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.01016cs.LGcs.AI
收藏

基于探索性采样的智能定位加速逆学习

Accelerating Inverse Learning via Intelligent Localization with Exploratory Sampling

Intuit AI Research(Intuit AI研究院);Walmart Global Tech(沃尔玛全球技术);Georgia Institute of Technology(佐治亚理工学院)

Jiaxin Zhang, Sirui Bi, Victor Fung

摘要:

在“人工智能科学”的范围内,解决逆问题是材料和药物发现中的一个长期挑战,其目标是确定给定一组理想属性的隐藏结构。最近提出了深度生成模型来解决逆问题,但这些模型目前使用昂贵的前向算子,并且难以精确定位精确解并充分探索参数空间而不会丢失解。在这项工作中,我们提出了一种新的方法(称为iPage),通过利用深度可逆模型的概率推理和通过快速梯度下降的确定性优化来加速逆学习过程。给定目标属性,学习的可逆模型提供了参数空间的后验;我们将这些后验样本识别为智能先验初始化,这使我们能够缩小搜索空间。然后,我们执行梯度下降来校准局部区域内的逆解。同时,对潜在空间进行空间填充采样,以更好地探索和捕获所有可能的解决方案。我们在三个基准任务和两个创建的数据集上评估了我们的方法,这些数据集具有来自量子化学和增材制造的实际应用,并发现我们的方法与几种最先进的基线方法相比具有更好的性能。iPage代码可在https://github.com/jxzhangjhu/MatDesINNe上获得。

英文摘要:

In the scope of "AI for Science", solving inverse problems is a longstanding challenge in materials and drug discovery, where the goal is to determine the hidden structures given a set of desirable properties. Deep generative models are recently proposed to solve inverse problems, but these currently use expensive forward operators and struggle in precisely localizing the exact solutions and fully exploring the parameter spaces without missing solutions. In this work, we propose a novel approach (called iPage) to accelerate the inverse learning process by leveraging probabilistic inference from deep invertible models and deterministic optimization via fast gradient descent. Given a target property, the learned invertible model provides a posterior over the parameter space; we identify these posterior samples as an intelligent prior initialization which enables us to narrow down the search space. We then perform gradient descent to calibrate the inverse solutions within a local region. Meanwhile, a space-filling sampling is imposed on the latent space to better explore and capture all possible solutions. We evaluate our approach on three benchmark tasks and two created datasets with real-world applications from quantum chemistry and additive manufacturing, and find our method achieves superior performance compared to several state-of-the-art baseline methods. The iPage code is available at https://github.com/jxzhangjhu/MatDesINNe.

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。

↑