In this work, we propose an efficient implementation of mixtures of experts distributional regression models which exploits robust estimation by using stochastic first-order optimization techniques with adaptive learning rate schedulers. We take advantage of the flexibility and scalability of neural network software and implement the proposed framework in mixdistreg, an R software package that allows for the definition of mixtures of many different families, estimation in high-dimensional and large sample size settings and robust optimization based on TensorFlow. Numerical experiments with simulated and real-world data applications show that optimization is as reliable as estimation via classical approaches in many different settings and that results may be obtained for complicated scenarios where classical approaches consistently fail.
翻译:在这项工作中,我们建议高效实施专家分布回归模型的混合物,这些模型利用与适应性学习速率调度器一起的随机第一阶优化技术进行稳健的估计。我们利用神经网络软件的灵活性和可扩缩性,并在混合数据中实施拟议框架,这是一个R软件包,可以定义许多不同家庭的混合物,在高维和大样本尺寸环境中进行估算,并在TensorFlow的基础上进行强有力的优化。模拟和真实世界数据应用的量化实验表明,在许多不同环境中,通过传统方法的估算,优化与传统方法的估算一样可靠,对于传统方法一贯失败的复杂情景,可以取得结果。