Mixture of Experts Distributional Regression: Implementation Using Robust Estimation With Adaptive First-Order Methods

MCML Authors

David Rügamer

Prof. Dr.

Principal Investigator

Statistics, Data Science and Machine Learning

Florian Pfisterer

Dr.

* Former Member

→ Group Bernd Bischl
 Statistical Learning and Data Science

Bernd Bischl

Prof. Dr.

Director

Statistical Learning and Data Science

Abstract

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.

article

---

Advances in Statistical Analysis

Nov. 2023.

---

Authors

D. Rügamer • F. Pfisterer • B. Bischl • B. Grün

---

Links

DOI

---

Research Area

 A1 | Statistical Foundations & Explainability