面向分布外推荐的条件可识别隐环境建模
Conditionally Identifiable Latent-Environment Modeling for Out-of-Distribution Recommendation
浏览论文内容
中文总结 AI 辅助
针对分布外推荐的偏好偏移问题，提出条件可识别隐环境推荐模型CILER，在三类偏移下的12项OOD排序指标上均实现提升。
中文摘要 AI 辅助
分布外（OOD）推荐易受隐环境引发的偏好偏移影响。现有方法可从记录的交互中推断隐状态，但隐环境的统计意义及其对偏好的影响仍未明确。我们将该任务形式化为条件可识别风险感知推荐（CI-RR），并提出条件可识别隐环境推荐（CILER）。CILER采用用户条件指数族建模隐环境，采用特征索引多项式指定其对偏好的影响方式，通过对推断出的环境分布边缘化项目概率进行预测。在充分变异、正确设定和解码器正则性条件下，CILER可识别环境敏感表示至指定等价类。我们进一步将部署超额对数风险约束为环境推断误差。受控研究测试了充分变异和模型设定的可观测后果。在三个数据集上的实验表明，CILER在共享支持下的特征、时间和地理偏移下，提升了全部12项OOD排序指标。
英文摘要
Out-of-distribution (OOD) recommendation is vulnerable to preference shifts induced by a latent environment. Existing methods can infer latent states from logged interactions, yet the statistical meaning of the latent environment and its effect on preference remain underdetermined. We formulate this task as conditionally identifiable risk-aware recommendation (CI-RR) and propose Conditionally Identifiable Latent-Environment Recommendation (CILER). CILER uses a user-conditioned exponential family to model the latent environment and a feature-indexed polynomial to specify how it changes preference. It predicts by marginalizing item probabilities over the inferred environment distribution. Under sufficient variation, correct specification, and decoder regularity, CILER identifies the environment-sensitive representation up to the stated equivalence class. We further bound excess deployment log-risk by environment-inference error. Controlled studies test the observable consequences of sufficient variation and model specification. Experiments on three datasets show that CILER improves all twelve OOD ranking metrics under feature, temporal, and geographical shifts within shared support.
发表机构
- Southern University of Science and Technology（南方科技大学）
- Massey University（梅西大学）
机构由 AI 辅助整理，请以论文原文为准。
