arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.08198cs.LGcs.AI
收藏

近似推理的随机顺序消息传递算法分析

Analysis of Random Sequential Message Passing Algorithms for Approximate Inference

Centre for Systems Modelling and Quantitative Biomedicine, University of Birmingham(伯明翰大学系统建模与定量生物医学中心)

Burak Çakmak, Yue M. Lu, Manfred Opper

摘要:

我们分析了一个随机顺序消息传递算法的动态近似推理与大高斯潜变量模型中的学生-教师的情况。为了对潜在变量之间的非平凡依赖关系进行建模,我们假设从旋转不变集合中提取随机协方差矩阵。此外,我们考虑一个模型不匹配的设置,其中教师模型和学生使用的模型可能是不同的。通过动力学泛函方法,我们得到了精确的动力学平均场方程表征推理算法的动力学。我们还推导出一系列的模型参数的顺序算法不收敛。该参数范围的边界与静态概率模型的副本对称分析的de Almeida无源(AT)稳定性条件相吻合。

英文摘要:

We analyze the dynamics of a random sequential message passing algorithm for approximate inference with large Gaussian latent variable models in a student-teacher scenario. To model nontrivial dependencies between the latent variables, we assume random covariance matrices drawn from rotation invariant ensembles. Moreover, we consider a model mismatching setting, where the teacher model and the one used by the student may be different. By means of dynamical functional approach, we obtain exact dynamical mean-field equations characterizing the dynamics of the inference algorithm. We also derive a range of model parameters for which the sequential algorithm does not converge. The boundary of this parameter range coincides with the de Almeida Thouless (AT) stability condition of the replica symmetric ansatz for the static probabilistic model.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑