用因果嵌入指导元学习，提升跨任务迁移的可靠性。
Bayesian Meta-Learning with Expert Feedback for Task-Shift Adaptation through Causal Embeddings
- 基于因果嵌入构建任务先验，避免错误关联导致的负迁移
- 在真实临床数据中实现跨疾病预测，显著减少负迁移
- 适合需要可靠跨领域迁移的医疗等高风险场景
元学习在分布内新任务上表现良好，但在分布外目标任务上常因源任务迁移引发负迁移而失效。本文提出一种基于因果嵌入的贝叶斯元学习方法，通过预计算的潜空间因果任务嵌入来条件化任务特定先验，使迁移基于机制相似性而非虚假相关性。该方法考虑实际部署中目标任务数据有限，依赖噪声较大的专家提供的源-目标任务因果相似性判断。理论分析表明，因果嵌入能控制先验不匹配，缓解任务偏移下的负迁移。实验显示，在受控模拟和大规模真实临床预测场景中，该方法均有效降低负迁移并提升分布外适应能力，其因果嵌入与潜在临床机制一致。
原文摘要 · Abstract (English)
Meta-learning methods perform well on new within-distribution tasks but often fail when adapting to out-of-distribution target tasks, where transfer from source tasks can induce negative transfer. We propose a causally-aware Bayesian meta-learning method, by conditioning task-specific priors on precomputed latent causal task embeddings, enabling transfer based on mechanistic similarity rather than spurious correlations. Our approach explicitly considers realistic deployment settings where access to target-task data is limited, and adaptation relies on noisy (expert-provided) pairwise judgments of causal similarity between source and target tasks. We provide a theoretical analysis showing that conditioning on causal embeddings controls prior mismatch and mitigates negative transfer under task shift. Empirically, we demonstrate reductions in negative transfer and improved out-of-distribution adaptation in both controlled simulations and a large-scale real-world clinical prediction setting for cross-disease transfer, where causal embeddings align with underlying clinical mechanisms.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。