I am currently a PhD student at the Otto-von-Guericke University Magdeburg (Germany) in the department of computer science.
My research focusses on anomaly detection in high-dimensional data with deep learning models, and more recently with neuro-symbolic methods, mostly on image, video and text data.
I also did some work on reproducibility of experiments involving deep learning models, as well as mining of academic literature.
I maintain pytorch-ood, a python library for out-of-distribution detection with deep neural networks.
In my spare time, I like to scape data from the web and mine it.
This hobby spawned some interesting projects with dedicated websites, such as sworm and extra-mining.
On the right, you can find a rather dated picture of me in the style of an image from Picasso’s blue period.
It was generated with Neural Style Transfer.
Our paper Deep learning-based harmonization and super-resolution of Landsat-8 and Sentinel-2 images, which is based on the masters thesis of my colleague Venkatesh Thirugnana Sambandham, has been published in the ISPRS Journal of Photogrammetry and Remote Sensing. This work is an extension of our previous workshop paper on transformers for satellite homogenization. In summary, we find that a simple UNet model is able to provide surprisingly good performance for the satellite homogenization task.
Our paper Out-of-Distribution Detction with Logical Reasoning has been accepted on the WACV 2024.
Abstract Machine Learning models often only generalize reliably to samples from the training distribution. Consequentially, detecting when input data is out-of-distribution (OOD) is crucial, especially in safety-critical applications. Current OOD detection methods, however, tend to be domain agnostic and often fail to incorporate valuable prior knowledge about the structure of the training distribution. To address this limitation, we introduce a novel, hybrid OOD detection algorithm that combines a deep learning-based perception system with a first-order logic-based knowledge representation.
My paper Towards Deep Anomaly Detection with Structured Knowledge Representations has been accepted on the Workshop on AI Safety Engineering at SafeComp.
Abstract Machine learning models tend to only make reliable predictions for inputs that are similar to the training data. Consequentially, anomaly detection, which can be used to detect unusual inputs, is critical for ensuring the safety of machine learning agents operating in open environments. In this work, we identify and discuss several limitations of current anomaly detection methods, such as their weak performance on tasks that require abstract reasoning, the inability to integrate background knowledge, and the opaqueness that undermines their trustworthiness in critical applications.
Did you know that the german parliament publishes protocols of all of its proceedings in PDF format? It is relatively straight-forward to download and parse them, so we can easily collect a dataset of transcripts of what seems to be every speech in the Bundestag since the second world war.
My original idea was to mine the speeches for word associations: some words will be associated with other words, based on the intended connotation, and this association might change over time as the connotations change.
I like to read tagesschau.de, so I wrote a script to scrape it in regular intervals. My original goal was to determine which articles stay on the front page the longest, which ones allow commenting (a feature that seems to have been disabled almost entirely since March 2020), and if articles are modified after the initial release (without mentioning this), because I sometimes feel that headlines change.
Dataset Creation Tagesschau.de provides a JSON API, so the “scraping” is relatively straight forward and can be done with just a few lines of code.
Our paper On Outlier Exposure with Generative Models has been accepted on the NeurIPS Machine Learning Safety Workshop.
Abstract While Outlier Exposure reliably increases the performance of Out-of-Distribution detectors, it requires a set of available outliers during training. In this paper, we propose Generative Outlier Exposure (GOE), which alleviates the need for available outliers by using generative models to sample synthetic outliers from low-density regions of the data distribution. The approach requires no modification of the generator, works on image and text data, and can be used with pre-trained models.
Our paper Social Work Research Map – ein niederschwelliger Zugang zu internationalen Publikationen der Sozialen Arbeit has been published in the journal Soziale Passagen.
There should have been a video here but your browser does not seem to support it.
Our abstract Towards Transformer-based Homogenization of Satellite Imagery for Landsat-8 and Sentinel-2 was accepted for presentation on the Transformers Workshop for Environmental Science.
In summary, we somewhat surprisingly found that transformers, a neural network architecture that achieves state-of-the-art results on most tasks it is applied to, does not outperform a vanilla U-Net model on our particular superresolution task.
What is Filter Visualization? Deep Neural Networks are often seen as a black-box: they map some input to some output, and we can make them do this surprisingly well. However, we usually have no idea how this mapping works.
Why would we want to do it? Visualizing filters can help us to get an understandin of what the neural network is doing. The method can also be used to identify filters that are not required for the model, because they are copies of other filters, or compute not valuable features at all.
Our Paper Multi-Class Hypersphere Anomaly Detection (MCHAD) has been accepted for presentation at the ICPR 2022. In summary, we propose a new loss function for learning neural networks that are able to detect anomalies in their inputs.
Poster for MCHAD (PDF).
How does it work? Omitting some details, the loss we propose has three different components, each of which we will explain in the following.
Intra-Class Variance We want the $f(x)$ of one class to cluster as tightly around a class center $\mu_y$ as possible.
Our paper PyTorch-OOD: A library for Out-of-Distribution Detection based on PyTorch has been presented at the CVPR 2022 Workshops. You can find the most recent version of the python source code on GitHub.
You can also install the package directly via pip:
pip install pytorch-ood The presentation slides are available here.
During the last weeks, I worked with some colleagues on a website that aims to improve access to social work literature.
There should have been a video here but your browser does not seem to support it.
Our companion paper On Challenging Aspects of Reproducibility in Deep Anomaly Detection has been accepted for presentation on the Fourth Workshop on Reproducible Research in Pattern Recognition (satellite event of ICPR 2022).
In it, we discuss aspects of reproducibility for our anomaly detection algorithm MCHAD, as well as anomaly detection with deep neural networks in general. In particular, we discussed the following challenges for the reproducibility:
Nondeterminism: conducting the same experiment with different random seeds might lead to significantly different outcomes.
Our Paper Addressing Randomness in Evaluation Protocols for Out-of-Distribution Detection has been accepted at the ICJAI 2021 Workshop for Artificial Intelligence for Anomalies and Novelties.
In summary, we investigated the following phenomenon: when you train neural networks several times, and then measure their performance on some task, there is a certain variance in the performance measurements, since the results of experiments may vary based on several factors (that are effectively controlled by the random seed).
Inspired by the “Spiegel-Mining” talk from David Kriesel, a friend of mine and a Prof. from the Hochschule Magdeburg scraped a german website that regularly publishes reviews of social work literature, and mined the resulting 18.000 articles, hoping for interesting insights.
In an attempt to visualize the discourse, we created several topic maps, like the one below, which you can find on the accompanying (german) website. The colours represent the gender of the authors of the review.
