Search Results for author: Stefan Lessmann

Found 24 papers, 8 papers with code

Leveraging Zero-Shot Prompting for Efficient Language Model Distillation

no code implementations23 Mar 2024 Lukas Vöge, Vincent Gurgul, Stefan Lessmann

This paper introduces a novel approach for efficiently distilling LLMs into smaller, application-specific models, significantly reducing operational costs and manual labor.

Language Modelling

The impact of heteroskedasticity on uplift modeling

no code implementations8 Dec 2023 Björn Bokelmann, Stefan Lessmann

In our research, we show that heteroskedastictity in the training data can cause a bias of the uplift model ranking: individuals with the highest treatment effects can get accumulated in large numbers at the bottom of the ranking.

Fair Models in Credit: Intersectional Discrimination and the Amplification of Inequity

no code implementations1 Aug 2023 Savina Kim, Stefan Lessmann, Galina Andreeva, Michael Rovatsos

Drawing from the intersectionality paradigm, the study examines intersectional horizontal inequities in credit access by gender, age, marital status, single parent status and number of children.

Decision Making Fairness

Multimodal Document Analytics for Banking Process Automation

no code implementations21 Jul 2023 Christopher Gerling, Stefan Lessmann

In sum, the paper contributes original empirical evidence on the effectiveness and efficiency of multi-model models for document processing in the banking business and offers practical guidance on how to unlock this potential in day-to-day operations.

token-classification Token Classification

The Deep Promotion Time Cure Model

1 code implementation19 May 2023 Victor Medina-Olivares, Stefan Lessmann, Nadja Klein

We propose a novel method for predicting time-to-event in the presence of cure fractions based on flexible survivals models integrated into a deep neural network framework.

Computational Efficiency

A Data-driven Case-based Reasoning in Bankruptcy Prediction

no code implementations2 Nov 2022 Wei Li, Wolfgang Karl Härdle, Stefan Lessmann

In addition, we delicately examine the explainability of the CBR system in the decision-making process of bankruptcy prediction.

Decision Making

Improving uplift model evaluation on RCT data

1 code implementation5 Oct 2022 Björn Bokelmann, Stefan Lessmann

We theoretically analyze the variance of uplift evaluation metrics and derive possible methods of variance reduction, which are based on statistical adjustment of the outcome.

Marketing

Forecasting Cryptocurrency Returns from Sentiment Signals: An Analysis of BERT Classifiers and Weak Supervision

no code implementations6 Apr 2022 Duygu Ider, Stefan Lessmann

Anticipating price developments in financial markets is a topic of continued interest in forecasting.

Leveraging Image-based Generative Adversarial Networks for Time Series Generation

no code implementations15 Dec 2021 Justin Hellermann, Stefan Lessmann

To leverage the advances of image-based generative models for the time series domain, we propose a two-dimensional image representation for time series, the Extended Intertemporal Return Plot (XIRP).

Disentanglement Time Series +2

Personalization in E-Grocery: Top-N versus Top-k Rankings

no code implementations30 May 2021 Franziska Scherpinski, Stefan Lessmann

To fill this gap and raise business performance, this paper introduces an RS with a personalized long ranking (top-N).

Recommendation Systems

Interpretable Multiple Treatment Revenue Uplift Modeling

no code implementations9 Jan 2021 Robin M. Gubela, Stefan Lessmann

Uplift models support a firm's decision-making by predicting the change of a customer's behavior due to a treatment.

Decision Making Marketing

Data driven value-at-risk forecasting using a SVR-GARCH-KDE hybrid

no code implementations15 Sep 2020 Marius Lux, Wolfgang Karl Härdle, Stefan Lessmann

The SVR-GARCH-KDE hybrid is compared to standard, exponential and threshold GARCH models coupled with different error distributions.

Density Estimation Management

Conditional Wasserstein GAN-based Oversampling of Tabular Data for Imbalanced Learning

1 code implementation20 Aug 2020 Justin Engelmann, Stefan Lessmann

Class imbalance is a common problem in supervised learning and impedes the predictive performance of classification models.

Classification General Classification

Targeting customers under response-dependent costs

2 code implementations13 Mar 2020 Johannes Haupt, Stefan Lessmann

This study provides a formal analysis of the customer targeting problem when the cost for a marketing action depends on the customer response and proposes a framework to estimate the decision variables for campaign profit optimization.

Causal Inference Marketing

Response Transformation and Profit Decomposition for Revenue Uplift Modeling

no code implementations20 Nov 2019 Robin M. Gubela, Stefan Lessmann, Szymon Jaroszewicz

The proposed methodology entails a transformation of the prediction target, customer-level revenues, that facilitates implementing a causal uplift model using standard machine learning algorithms.

BIG-bench Machine Learning Decision Making +1

Affordable Uplift: Supervised Randomization in Controlled Experiments

1 code implementation1 Oct 2019 Johannes Haupt, Daniel Jacob, Robin M. Gubela, Stefan Lessmann

To increase the cost-efficiency of experimentation and facilitate frequent data collection and model training, we introduce supervised randomization.

Decision Making Marketing

Churn Prediction with Sequential Data and Deep Neural Networks. A Comparative Analysis

1 code implementation24 Sep 2019 C. Gary Mena, Arno De Caigny, Kristof Coussement, Koen W. De Bock, Stefan Lessmann

Off-the-shelf machine learning algorithms for prediction such as regularized logistic regression cannot exploit the information of time-varying features without previously using an aggregation procedure of such sequential data.

regression

Can Deep Learning Predict Risky Retail Investors? A Case Study in Financial Risk Behavior Forecasting

no code implementations14 Dec 2018 Yaodong Yang, Alisa Kolesnikova, Stefan Lessmann, Tiejun Ma, Ming-Chien Sung, Johnnie E. V. Johnson

The results of employing a deep network for operational risk forecasting confirm the feature learning capability of deep learning, provide guidance on designing a suitable network architecture and demonstrate the superiority of deep learning over machine learning and rule-based benchmarks.

BIG-bench Machine Learning Feature Engineering +1

Health Analytics: a systematic review of approaches to detect phenotype cohorts using electronic health records

no code implementations24 Jul 2017 Norman Hiob, Stefan Lessmann

The paper presents a systematic review of state-of-the-art approaches to identify patient cohorts using electronic health records.

BIG-bench Machine Learning

Cannot find the paper you are looking for? You can Submit a new open access paper.