This page showcases academic research I have participated in. That is, projects that take place in an academic community with the intent of publishing. It is ordered from newest to oldest publication date.

Publications

📝 CURED: Creating, Understanding, and Repairing Errors Demonstrator

To summarize and make the work on errors and data quality accessible to a wider audience, I built a demonstrator and wrote an accompanying paper, CURED: Creating, Understanding, and Repairing Errors Demonstrator while at Calgo-Lab. The paper is available as an arXiv preprint.

Abstract:

Detecting and cleaning errors in tabular data is a prerequisite for data intense software applications. Recent research at the intersection of Machine Learning (ML) and Database Management Systems (DBMS) highlights the potential of statistical learning algorithms for error detection and cleaning. This paper combines our recent work on MLbased data cleaning and error models in a unified demonstrator. The web application allows users to upload tabular data, perturb the data with realistic data dependent errors and use modern ML methods to clean and understand error mechanisms in data. Our demonstrator1 helps to bridge the gap between theoretical advancements and intuitive practical insights in the context of error models and data cleaning algorithms for tabular data. The demonstrator is available at https://cured.demo.calgo-lab.de/.

Contributions:

I built the demonstrator and wrote the paper.

📝 Anonymized but Useful Synthetic Tabular Health Data for AI based Fall Risk Assessment

I assisted in writing code for the Synthius Python package which was used in the creation of the synthetic dataset in the paper Anonymized but Useful Synthetic Tabular Health Data for AI based Fall Risk Assessment. It was accepted and published in Nature Scientific Data.

Abstract:

Artificial Intelligence (AI) bears potential for improving health care, but this depends on the availability of open-access, realistic, and useful data. To facilitate AI model development in health care we release SynTabFall, a novel synthetic dataset for fall risk assessment. With a total of 745,380 samples and 44 attributes such as demographics, diseases, mobility and cognition related risk factors, this tabular dataset allows for training fall risk prediction models without access to the original patient data. Models trained on our synthetic dataset can reach predictive performance scores in fall risk assessment which are on par with models trained on real data. To support others in sharing data we also describe a process that was developed over multiple years in one of Germany’s largest hospitals in close collaboration between data protection officers, health care staff, informaticians and AI engineers. The proposed data sharing approach combines established methods for anonymization and modern generative AI (genAI) methods for synthesizing tabular data and allows for sharing health care data responsibly without sacrificing its utility. We release the synthetic fall risk dataset along with the software developed for synthetic data generation and evaluation.

Contributions:

I wrote code for Synthius including an API, documentation, and a getting started guide.

📝 Probabalistic Neural Network Approach to Determining Parameters of Eclipsing Binaries

I did research on developing a deep learning approach to determining the parameters of eclipsing binary star systems. The paper Probabilistic Neural Network Approach to Determining Parameters of Eclipsing Binaries was accepted and published in The Astronomical Journal.

Abstract:

Eclipsing binaries provide one of the most direct mechanisms for measuring stellar properties such as mass and radius, but historically, determining these properties has been non-trivial and computationally prohibitive. As such, only a small fraction of all eclipsing binaries for which data have been available have been fully characterized. To improve computational efficiency, we construct an uncertainty-aware neural network which can ingest phase-folded light curves in any of 50 commonly used passbands, combined with phase-folded radial velocity measurements for both primary and secondary, as well as fluxes across the spectral energy distribution to predict stellar and orbital parameters of eclipsing binaries. The model was trained to be agnostic to the presence of third light, spots (both cool and hot), and incomplete data. As the model is operating in a probabilistic framework, it is also capable of outputting uncertainties in all of the parameters. The model was trained on synthetic data, and applied to a set of ∼200 previously solved real eclipsing binaries to demonstrate its performance. The model is capable of determining masses and radii of eclipsing binaries with precision of ≲20% and Teff with precision of ∼500 K in only a fraction of the time it takes the more traditional solvers. Although the resulting uncertainties are larger than what is possible to produce using more boutique analysis of individual stars, in the era of large photometric surveys, this approach allows to identify the most interesting systems, and it provides a starting point of the distributions in all of the parameters that these solvers could improve upon.

Contributions:

I worked on prototyping the pytorch bayesian neural network as a part of undergraduate research at Western Washington University.

📝 MechDetect

I ran the experiments for (and co-authored) this paper: MechDetect: Detecting Data-Dependent Errors. It was accepted at the IEEE conference Data Science and Intelligent Systems 2025 and published in the proceedings.

Abstract:

Data quality monitoring is a core challenge in modern information processing systems. While many approaches to detect data errors or shifts have been proposed, few studies investigate the mechanisms governing error generation. We argue that knowing how errors were generated can be key to tracing and fixing them. In this study, we build on existing work in the statistics literature on missing values and propose MechDetect, a simple algorithm to investigate error generation mechanisms. Given a tabular data set and a corresponding error mask, the algorithm estimates whether or not the errors depend on the data using machine learning models. Our work extends established approaches to detect mechanisms underlying missing values and can be readily applied to other error types, provided that an error mask is available. We demonstrate the effectiveness of MechDetect in experiments on established benchmark datasets.

Contributions:

I ran the experiments, produced some of the scientific visualizations, and wrote the prose.

📝 Towards realistic error models for tabular data

I assisted with the revisions of (and co-authored) the paper: Towards Realistic Error Models for Tabular Data. It has been accepted and published in the Journal of Data and Information Quality of the ACM.

Abstract:

Errors in data are a key challenge in modern data management and processing systems. Monitoring and mitigating risks associated with errors in data transformations and downstream applications, such as Machine Learning (ML) model training, requires a profound understanding of error generation and impact of errors on data pipelines. Unfortunately, scientific progress in the field is facing two main challenges: For one, research on data errors often does not adhere to the FAIR (Findable, Accessible, Interoperable, and Reusable) principles, which impedes reproducibility and comparisons. Second, existing data error models are oversimplified and fail to capture the complex statistical dependencies underlying the types and distributions of errors observed in real-world data. Building on prior work in the database management systems and statistics literature, we extend the theory on missing values to encompass a broader range of errors in tables and provide an overview of relevant error types. Combining error sampling mechanisms often observed in real data with a comprehensive categorization of errors, we introduce a latent factor model for tabular data errors that is simple to implement and can effectively model realistic error dependencies. Error sampling is decoupled from error types, which allows for simple extensions with more error types or sampling mechanisms. Using established benchmarks, we evaluate our model in two application scenarios, data cleaning and tabular ML tasks. In a comprehensive suite of experiments we demonstrate the impact of realistic error models on data cleaning benchmarks. Our results also show that a simple generative error model captures a wide range of error mechanisms and offers a convenient formalization of data perturbations to improve the generalizability, robustness and reproducibility of data cleaning research.

Contributions:

I ran experiments to determine the scalability of the library with respect to memory and runtime. I also ran experiments to determine how different error models affect the downstream performance of machine learning models. Finally, I aided in the presentation of the results via prose checking and producing scientific visualizations.

📖 Editing/ Errata Addition of Machine Learning Kurz & Gut

  • I read the book Machine Learning Kurz und Gut in German and pointed out a few errors to the author.
  • Shows some proficiency in German written communication.
  • Github

Presentations

🗣️ PiMUC 2024 Statistics Research Presentation (WWU)

  • In this project, I present an example application of a novel statistical test to reaction times data.
  • This was worked on under the supervision of Professor Kimihiro Noguchi of the WWU Mathematics department.
  • We obtained funding from the WWU Mathematics department to travel to the 2024 Pacific inland Undergraduate Mathematics Conference (PiMUC) to present.
  • Presentation

🗣️ PiMUC 2023 Statistics Research Presentation (WWU)

  • In this project, I present a statistical test developed by Professor Kimihiro Noguchi of the WWU Mathematics department, to which I contributed by running simulation study code on the compute cluster and doing data visualization.
  • This was worked on under the supervision of Professor Kimihiro Noguchi of the WWU Mathematics department.
  • We obtained funding from the WWU Mathematics department to travel to the 2023 Pacific inland Undergraduate Mathematics Conference (PiMUC) to present.
  • Presentation

Posters

🧮 Schorfheide Summer School 2026 Poster (BHT)

  • In this poster, I presented some preliminary results from my master’s thesis.
  • The poster concerns spatiotemporal modeling of groundwater with the goal of making forecasts of groundwater levels at sites with no historical data.
  • Done in association with Calgo Lab under supervision of Professor Felix Biessmann and PhD candidate Vipin Singh
  • Poster

🧮 Scholar’s Week 2024 Poster (WWU)

  • In this poster, I presented results from my senior research project with my team members.
  • This project was concerned with the prediction of various stellar parameters given satellite data using deep learning.
  • Done in the HutchResearch group at WWU.
  • Poster

🧮 Joint Mathematics Meetings 2024 Poster (WWU)

  • In this poster, I presented research regarding the simulation study aspect of the development of a novel class of statistical tests at the 2024 Joint Mathematics Meetings in San Francisco.
  • This was done under supervision of Professor Kimihiro Noguchi of the WWU Mathematics department.
  • We obtained funding from the WWU Mathematics department to travel to and stay in San Francisco for the 2024 Joint Mathematics Meetings.
  • Poster

🧮 Society of Industrial and Applied Mathematicians 2023 Biannual Meeting Poster (WWU)

  • In this poster, I presented on the simulation study component of the development of a novel class of non-parametric statistical tests developed by Professor Kimihiro Noguchi in the WWU Mathematics Department.
  • The SIAM 2023 Biannual Meeting was hosted at WWU.
  • Poster

Organizations