Deep Conditional Transformation Models

Baumann, Philipp F. M.; Hothorn, Torsten; Rügamer, David

doi:10.1007/978-3-030-86523-8_1

Cited by 9 publications

(8 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To allow for a more nuanced view on uncertainty, we follow Hüllermeier and Waegeman (2021) and distinguish between two sources of uncertainty for any machine learning task: while epistemic uncertainty accounts for uncertainty in the model that can generally be reduced given sufficient data, this paper is concerned with aleatoric uncertainty, which reflects the randomness inherent in observations. This type of uncertainty can be captured by modelling a conditional outcome probability distribution that is dependent on features (Baumann et al, 2021).…”

Section: Dgbm: Distributional Gradient Boosting Machinesmentioning

confidence: 99%

“…that is learnt from the data (Baumann et al, 2021). However, instead of a transformation from z to y, Transformation Models define an inverse flow h(y) = z (Rügamer et al, 2022).…”

Section: B Nfboost: Normalizing Flow Boostingmentioning

confidence: 99%

“…The authors have successfully applied neural Bernstein Flow models for short-term load forecasting, UCI datasets, as well as for approximating posteriors in Variational Bayes inference. Furthermore, Rügamer et al (2022); Baumann et al (2021) show stateof-the-art performance of Conditional Transformation based neural networks applied to time series, as well as UCI datasets.…”

Section: Related Researchmentioning

confidence: 99%

See 2 more Smart Citations

Distributional Gradient Boosting Machines

März¹,

Kneib²

2022

Preprint

View full text Add to dashboard Cite

We present a unified probabilistic gradient boosting framework for regression tasks that models and predicts the entire conditional distribution of a univariate response variable as a function of covariates. Our likelihood-based approach allows us to either model all conditional moments of a parametric distribution, or to approximate the conditional cumulative distribution function via Normalizing Flows. As underlying computational backbones, our framework is based on XGBoost and LightGBM. Modelling and predicting the entire conditional distribution greatly enhances existing tree-based gradient boosting implementations, as it allows to create probabilistic forecasts from which prediction intervals and quantiles of interest can be derived. Empirical results show that our framework achieves state-of-the-art forecast accuracy.

show abstract

Section: Dgbm: Distributional Gradient Boosting Machinesmentioning

confidence: 99%

“…that is learnt from the data (Baumann et al, 2021). However, instead of a transformation from z to y, Transformation Models define an inverse flow h(y) = z (Rügamer et al, 2022).…”

Section: B Nfboost: Normalizing Flow Boostingmentioning

confidence: 99%

Section: Related Researchmentioning

confidence: 99%

See 1 more Smart Citation

Distributional Gradient Boosting Machines

März¹,

Kneib²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Recent approaches (Rügamer et al, 2020) suggest to fit structured regression models as (part of of) a neural network. This can result in a better space complexity in situations with many data points and allows for more flexibility in the additive predictors of models beyond those of classical GAMs (see, e.g., Baumann et al, 2021;Kopper et al, 2021).…”

Section: Related Literaturementioning

confidence: 99%

Additive Higher-Order Factorization Machines

Rügamer¹

2022

Preprint

Self Cite

View full text Add to dashboard Cite

In the age of big data and interpretable machine learning, approaches need to work at scale and at the same time allow for a clear mathematical understanding of the method's inner workings. While there exist inherently interpretable semi-parametric regression techniques for large-scale applications to account for non-linearity in the data, their model complexity is still often restricted. One of the main limitations are missing interactions in these models, which are not included for the sake of better interpretability, but also due to untenable computational costs. To address this shortcoming, we derive a scalable high-order tensor product spline model using a factorization approach. Our method allows to include all (higher-order) interactions of non-linear feature effects while having computational costs proportional to a model without interactions. We prove both theoretically and empirically that our methods scales notably better than existing approaches, derive meaningful penalization schemes and also discuss further theoretical aspects. We finally investigate predictive and estimation performance both with synthetic and real data.

show abstract

“…A recent trend is the combination of neural networks with statistical regression models in various ways [41,42,43,44,45,46,47]. In this work, we make use of semi-structured deep distributional regression [SDDR,48].…”

Section: Semi-structured Deep Distributional Regressionmentioning

confidence: 99%

Combining Graph Neural Networks and Spatio-temporal Disease Models to Predict COVID-19 Cases in Germany

Fritz,

Dorigatti,

Rügamer

2021

Preprint

Self Cite

View full text Add to dashboard Cite

During 2020, the infection rate of COVID-19 has been investigated by many scholars from different research fields. In this context, reliable and interpretable forecasts of disease incidents are a vital tool for policymakers to manage healthcare resources. Several experts have called for the necessity to account for human mobility to explain the spread of COVID-19. Existing approaches are often applying standard models of the respective research field. This habit, however, often comes along with certain restrictions. For instance, most statistical or epidemiological models cannot directly incorporate unstructured data sources, including relational data that may encode human mobility. In contrast, machine learning approaches may yield better predictions by exploiting these data structures, yet lack intuitive interpretability as they are often categorized as black-box models. We propose a trade-off between both research directions and present a multimodal learning approach that combines the advantages of statistical regression and machine learning models for predicting local COVID-19 cases in Germany. This novel approach enables the use of a richer collection of data types, including mobility flows and colocation probabilities, and yields the lowest MSE scores throughout our observational period in our benchmark study. The results corroborate the necessity of including mobility data and showcase the flexibility and interpretability of our approach.

show abstract

Deep Conditional Transformation Models

Cited by 9 publications

References 24 publications

Distributional Gradient Boosting Machines

Distributional Gradient Boosting Machines

Additive Higher-Order Factorization Machines

Combining Graph Neural Networks and Spatio-temporal Disease Models to Predict COVID-19 Cases in Germany

Contact Info

Product

Resources

About