arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.02744cs.LG
收藏

SALSA:语义感知潜空间自编码器

SALSA: Semantically-Aware Latent Space Autoencoder

University of North Carolina(北卡罗来纳大学)

Kathryn E. Kirchoff, Travis Maxfield, Alexander Tropsha, Shawn M. Gomez

AI总结:

针对仅用SMILES训练的自编码器潜空间不尊重分子结构相似性的问题,本文提出带监督对比损失的SALSA,使仅差一次图编辑的分子编码接近,从而提升潜空间质量。

摘要:

在药物发现的深度学习中,化学数据通常被表示为简化分子输入线性条目系统(SMILES)序列,这使得自然语言处理方法可以直接应用,序列到序列自编码器便是其中之一。然而,我们观察到,仅基于SMILES训练自编码器不足以学到具有语义意义的分子表示;这里的语义由分子之间的结构(图到图)相似性定义。我们通过示例证明,自编码器可能会将结构相似的分子映射到相距较远的编码,从而产生不一致的潜空间,不能反映分子之间的结构相似性。为解决这一缺陷,我们提出语义感知潜空间自编码器(Semantically-Aware Latent Space Autoencoder,SALSA),它是一种经过对比任务改造的transformer自编码器,专门用于学习分子之间的图到图相似性。形式上,对比目标是将结构相似的分子(仅相差一次图编辑)映射到潜空间中彼此接近的编码。为此,我们构建了一个由多组结构相似分子组成的新型数据集,并采用能够整合完整正样本集合的监督对比损失。我们将SALSA与其消融版本进行比较,并通过实验表明,由重建和对比任务组成的训练目标能够产生更高质量的潜空间,该空间更具1)结构感知性、2)语义连续性和3)性质感知性。

英文摘要:

In deep learning for drug discovery, chemical data are often represented as simplified molecular-input line-entry system (SMILES) sequences which allow for straightforward implementation of natural language processing methodologies, one being the sequence-to-sequence autoencoder. However, we observe that training an autoencoder solely on SMILES is insufficient to learn molecular representations that are semantically meaningful, where semantics are defined by the structural (graph-to-graph) similarities between molecules. We demonstrate by example that autoencoders may map structurally similar molecules to distant codes, resulting in an incoherent latent space that does not respect the structural similarities between molecules. To address this shortcoming we propose Semantically-Aware Latent Space Autoencoder (SALSA), a transformer-autoencoder modified with a contrastive task, tailored specifically to learn graph-to-graph similarity between molecules. Formally, the contrastive objective is to map structurally similar molecules (separated by a single graph edit) to nearby codes in the latent space. To accomplish this, we generate a novel dataset comprised of sets of structurally similar molecules and opt for a supervised contrastive loss that is able to incorporate full sets of positive samples. We compare SALSA to its ablated counterparts, and show empirically that the composed training objective (reconstruction and contrastive task) leads to a higher quality latent space that is more 1) structurally-aware, 2) semantically continuous, and 3) property-aware.

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。

↑