让模型学会发现因果关系，用少量样本就能推理新任务。
Causal-Symbolic Meta-Learning (CSML): Inducing Causal World Models for Few-Shot Generalization
- 用符号化表示+可微因果推断，自动发现任务背后的因果图
- 在仅1~3个例子下完成干预与反事实推理，性能远超现有方法
- 适合需要可靠推理的少样本场景，如医疗、机器人决策
现代深度学习模型擅长模式识别，但依赖表面相关性，泛化能力差且需海量数据。我们认为，类人智能的关键在于理解因果机制，实现鲁棒、高效的学习。本文提出因果-符号元学习（CSML）框架，通过元学习共享的因果世界模型，从任务分布中推断潜在因果结构。该框架包含三个模块：感知模块将原始输入映射为解耦符号表示；可微因果推断模块发现控制这些符号的底层因果图；图推理模块利用该图进行预测。通过在任务分布上元学习因果模型，CSML可在仅1~3个示例下快速适应新任务，包括涉及干预和反事实推理的任务。我们构建了基于物理的新基准CausalWorld以测试此能力。实验表明，CSML显著优于现有元学习与神经符号基线，尤其在真正因果推理任务上表现突出。
原文摘要 · Abstract (English)
Modern deep learning models excel at pattern recognition but remain fundamentally limited by their reliance on spurious correlations, leading to poor generalization and a demand for massive datasets. We argue that a key ingredient for human-like intelligence-robust, sample-efficient learning-stems from an understanding of causal mechanisms. In this work, we introduce Causal-Symbolic Meta-Learning (CSML), a novel framework that learns to infer the latent causal structure of a task distribution. CSML comprises three key modules: a perception module that maps raw inputs to disentangled symbolic representations; a differentiable causal induction module that discovers the underlying causal graph governing these symbols and a graph-based reasoning module that leverages this graph to make predictions. By meta-learning a shared causal world model across a distribution of tasks, CSML can rapidly adapt to novel tasks, including those requiring reasoning about interventions and counterfactuals, from only a handful of examples. We introduce CausalWorld, a new physics-based benchmark designed to test these capabilities. Our experiments show that CSML dramatically outperforms state-of-the-art meta-learning and neuro-symbolic baselines, particularly on tasks demanding true causal inference.
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。