How Inverse Conditional Flows Can Serve as a Substitute for Distributional Regression
Authors:
Lucas Kook,
Chris Kolb,
Philipp Schiele,
Daniel Dold,
Marcel Arpogaus,
Cornelius Fritz,
Philipp F. Baumann,
Philipp Kopper,
Tobias Pielok,
Emilio Dorigatti,
David Rügamer
Abstract:
Neural network representations of simple models, such as linear regression, are being studied increasingly to better understand the underlying principles of deep learning algorithms. However, neural representations of distributional regression models, such as the Cox model, have received little attention so far. We close this gap by proposing a framework for distributional regression using inverse…
▽ More
Neural network representations of simple models, such as linear regression, are being studied increasingly to better understand the underlying principles of deep learning algorithms. However, neural representations of distributional regression models, such as the Cox model, have received little attention so far. We close this gap by proposing a framework for distributional regression using inverse flow transformations (DRIFT), which includes neural representations of the aforementioned models. We empirically demonstrate that the neural representations of models in DRIFT can serve as a substitute for their classical statistical counterparts in several applications involving continuous, ordered, time-series, and survival outcomes. We confirm that models in DRIFT empirically match the performance of several statistical methods in terms of estimation of partial effects, prediction, and aleatoric uncertainty quantification. DRIFT covers both interpretable statistical models and flexible neural networks opening up new avenues in both statistical modeling and deep learning.
△ Less
Submitted 13 May, 2024; v1 submitted 8 May, 2024;
originally announced May 2024.
Short-Term Density Forecasting of Low-Voltage Load using Bernstein-Polynomial Normalizing Flows
Authors:
Marcel Arpogaus,
Marcus Voss,
Beate Sick,
Mark Nigge-Uricher,
Oliver Dürr
Abstract:
The transition to a fully renewable energy grid requires better forecasting of demand at the low-voltage level to increase efficiency and ensure reliable control. However, high fluctuations and increasing electrification cause huge forecast variability, not reflected in traditional point estimates. Probabilistic load forecasts take future uncertainties into account and thus allow more informed dec…
▽ More
The transition to a fully renewable energy grid requires better forecasting of demand at the low-voltage level to increase efficiency and ensure reliable control. However, high fluctuations and increasing electrification cause huge forecast variability, not reflected in traditional point estimates. Probabilistic load forecasts take future uncertainties into account and thus allow more informed decision-making for the planning and operation of low-carbon energy systems. We propose an approach for flexible conditional density forecasting of short-term load based on Bernstein polynomial normalizing flows, where a neural network controls the parameters of the flow. In an empirical study with 363 smart meter customers, our density predictions compare favorably against Gaussian and Gaussian mixture densities. Also, they outperform a non-parametric approach based on the pinball loss for 24h-ahead load forecasting for two different neural network architectures.
△ Less
Submitted 15 June, 2023; v1 submitted 29 April, 2022;
originally announced April 2022.
