arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2003.02359stat.MLcs.LGmath.DSphysics.data-an
收藏

贝叶斯系统ID:参数、模型和测量不确定性的最佳管理

Bayesian System ID: Optimal management of parameter, model, and measurement uncertainty

University of Michigan(密歇根大学)

Nicholas Galioto, Alex Gorodetsky

摘要:

我们评估的系统识别(ID)的概率公式的鲁棒性稀疏,嘈杂,和间接的数据。具体地说,我们比较了来自学习问题的贝叶斯后验的未来系统行为的估计量与几种常用的最小二乘,我们的比较表明,与传统方法的目标函数表面相比,对数后验具有改进的几何特性,传统方法包括离散时间步长的差分约束最小二乘和最小二乘重建,如动态模式分解(DMD)。这些属性使它对新数据更敏感,受多个最小值的影响更小-总体上产生了一种更鲁棒的方法。我们的理论结果表明,最小二乘和正则化最小二乘方法,如动态模式分解和稀疏识别的非线性动力学(SINDy),可以从概率公式假设无噪声测量。我们还分析了高斯滤波器为基础的近似边缘马尔可夫链蒙特卡罗计划,我们用来获得线性和非线性问题的贝叶斯后验的计算复杂性。然后,我们经验性地证明,获得参数动态的边际后验并通过提取最佳估计量(例如,平均值、中位数、众数)比上述方法产生了数量级的改进。我们将这种性能归因于贝叶斯方法捕获参数,模型和测量不确定性,而其他方法通常忽略至少一种类型的不确定性。

英文摘要:

We evaluate the robustness of a probabilistic formulation of system identification (ID) to sparse, noisy, and indirect data. Specifically, we compare estimators of future system behavior derived from the Bayesian posterior of a learning problem to several commonly used least squares-based optimization objectives used in system ID. Our comparisons indicate that the log posterior has improved geometric properties compared with the objective function surfaces of traditional methods that include differentially constrained least squares and least squares reconstructions of discrete time steppers like dynamic mode decomposition (DMD). These properties allow it to be both more sensitive to new data and less affected by multiple minima --- overall yielding a more robust approach. Our theoretical results indicate that least squares and regularized least squares methods like dynamic mode decomposition and sparse identification of nonlinear dynamics (SINDy) can be derived from the probabilistic formulation by assuming noiseless measurements. We also analyze the computational complexity of a Gaussian filter-based approximate marginal Markov Chain Monte Carlo scheme that we use to obtain the Bayesian posterior for both linear and nonlinear problems. We then empirically demonstrate that obtaining the marginal posterior of the parameter dynamics and making predictions by extracting optimal estimators (e.g., mean, median, mode) yields orders of magnitude improvement over the aforementioned approaches. We attribute this performance to the fact that the Bayesian approach captures parameter, model, and measurement uncertainties, whereas the other methods typically neglect at least one type of uncertainty.

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。

↑