arXiv cs.AI· Yuchen Yan, Yatao Bian·· 3 天前AI 评分22
PerturbEnergy：用条件能量先验学习扰动特异性表征预测细胞响应
Learning Perturbation-Specific Representations through Conditional Energy Prior for Cellular Response Prediction
AI 导读
研究者提出 PerturbEnergy，通过在扰动表征上引入条件能量先验，对与给定扰动兼容的表征赋予更低能量，从而将潜空间显式偏向扰动偏好区域。该方法可集成进自编码器架构，其 ELBO 与完整自编码器的 ELBO 等价，并被证明能提升表征与扰动间条件互信息的下界。
正文 · 原文
Abstract:AI-based cellular perturbation prediction is essential to accelerate drug discovery and therapeutic identification. However, perturbation-specific signals are generally obscured by cellular complexity, making it difficult for existing methods to correctly capture perturbation effects. To address this problem, we propose PerturbEnergy, which introduces a conditional energy-based prior over perturbation representations. It assigns lower energy (higher probability) to representations compatible with the given perturbation and higher energy to others, thereby explicitly tilting the latent space toward perturbation-preferred regions. Besides, PerturbEnergy can be integrated into autoencoder architectures as we find its Evidence Lower Bound (ELBO) is equivalent to the ELBO of the complete autoencoder. We further prove that optimizing PerturbEnergy increases the lower bound on the conditional mutual information between representation and perturbation. Extensive experiments on four benchmark datasets demonstrate the effectiveness of our PerturbEnergy: it improves perturbation recognition especially in latent space with improvement of up to 83.2\%; It also achieves state-of-the-art performance for predicting top-50/100/200 differentially expressed gene(DEG) response and remaining competitive on highly variable gene(HVG) prediction. The code is provided: this https URL
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.18456 [cs.LG] |
| (or arXiv:2610.18456v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.18456 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yuchen Yan [view email]
[v1]
Tue, 13 Oct 2026 14:24:25 UTC (581 KB)
来源：arXiv cs.AI · arxiv.org