一个神经网络搞定不同模型的快速贝叶斯推断，无需重训。
Amortized Factor Inference Networks for Posterior Inference
- 用模块化结构将模型描述和观测映射为后验参数
- 单个模型在测试时计算量减少100到10000倍，精度接近最优方法
- 适合需要快速适应新模型的科研与工程场景
摊销推断能实现快速的贝叶斯推断，但现有方法仅适用于固定模型。当面对未见模型时，通常需重新训练或昂贵的测试阶段微调。本文提出：能否构建一个能跨不同先验、似然函数和维度泛化的单一推断网络？我们引入摊销因子推断网络（AFINs），一种基于维度无关模块的编码-合并-解码架构，可将模型规格与观测数据映射为变分后验参数。实验表明，单个训练好的AFIN在后验精度上媲美NUTS和多种变分推断方法，而测试阶段计算量降低2至4个数量级。代码已开源：https://github.com/joohwanko/AFINs。
原文摘要 · Abstract (English)
Amortized inference promises fast test-time Bayesian inference, but existing methods are inherently tied to fixed models. Extending amortization to unseen models typically requires retraining or costly test-time finetuning. In this paper, we ask: is it possible to build a single inference network capable of generalizing across varying priors, likelihoods, and dimensionality? We introduce Amortized Factor Inference Networks (AFINs), a family of encode-merge-decode inference networks built on dimension-independent modules that map a model specification and its observations to the parameters of a variational posterior. Experimentally, a single trained AFIN achieves posterior accuracy comparable to NUTS and several variational inference methods, while requiring 2 to 4 orders of magnitude less test-time compute. Code is available at https://github.com/joohwanko/AFINs.
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。