about | Markus Thill

Hi there! I’m Markus, a Data Scientist at Atruvia AG. With a background in computer engineering (B.Sc.) and Automation & IT (M.Eng.), I’m passionate about the intersection of machine learning, mathematics, and programming. In my professional role, I delve into data to uncover insights and drive innovation.

I have a particular interest in machine learning approaches, especially in the fields of time series analysis, anomaly detection, Reinforcement Learning (RL) (such as in board games), Deep Learning (DL), and incremental learning procedures. Recently, I received my doctoral certificate from Leiden University for my work titled “Machine Learning and Deep Learning Approaches for Multivariate Time Series Prediction and Anomaly Detection”. This marked a final milestone in my academic journey, and since then, I’ve transitioned into the industry. During my tenure as a research associate at TH Köln, I presented a total of eight peer-reviewed publications at various international research conferences.

Outside of work, I find joy in simple pleasures like flying drones and occasionally solving small mathematical riddles on projecteuler.net. Continuously learning and exploring new concepts in data science and mathematics keeps me motivated and engaged.

news

Jan 15, 2016	A simple inline announcement with Markdown emoji!
Nov 07, 2015	A long announcement with details
Oct 22, 2015	A simple inline announcement.

latest posts

Apr 26, 2024	Solving Peg Solitaire with efficient Bit-Board Representations
Mar 31, 2024	Choosing a Voltage Divider Resistor for a Light Dependent Resistor
Mar 29, 2024	Deriving a Closed-Form Solution of the Fibonacci Sequence

selected publications

ASoC

Temporal convolutional autoencoder for unsupervised anomaly detection in time series

Markus Thill , Wolfgang Konen , Hao Wang , and 1 more author

Applied Soft Computing (ASoC), 2021

Abs

Learning temporal patterns in time series remains a challenging task up until today. Particularly for anomaly detection in time series, it is essential to learn the underlying structure of a system’s normal behavior. Periodic or quasiperiodic signals with complex temporal patterns make the problem even more challenging: Anomalies may be a hard-to-detect deviation from the normal recurring pattern. In this paper, we present TCN-AE, a temporal convolutional network autoencoder based on dilated convolutions. Contrary to many other anomaly detection algorithms, TCN-AE is trained in an unsupervised manner. The algorithm demonstrates its efficacy on a comprehensive real-world anomaly benchmark comprising electrocardiogram (ECG) recordings of patients with cardiac arrhythmia. TCN-AE significantly outperforms several other unsupervised state-of-the-art anomaly detection algorithms. Moreover, we investigate the contribution of the individual enhancements and show that each new ingredient improves the overall performance on the investigated benchmark.
ITAT

Anomaly Detection in Electrocardiogram Readings with Stacked LSTM Networks

Markus Thill , Sina Däubner , Wolfgang Konen , and 1 more author

In Proc. 19th Conference Information Technologies - Applications and Theory (ITAT 2019) , 2019

Best Paper Award

Abs

Real-world anomaly detection for time series is still a challenging task. This is especially true for periodic or quasi-periodic time series since automated approaches have to learn long-term correlations before they are able to detect anomalies. Electrocardiography (ECG) time series, a prominent real-world example of quasi-periodic signals, are investigated in this work. Anomaly detection algorithms often have the additional goal to identify anomalies in an unsupervised manner. In this paper we present an unsupervised time series anomaly detection algorithm. It learns with recurrent Long Short-Term Memory (LSTM) networks to predict the normal time series behavior. The prediction error on several prediction horizons is used to build a statistical model of normal behavior. We propose new methods that are essential for a successful model-building process and for a high signal-to-noise-ratio. We apply our method to the well-known MIT-BIH ECG data set and present first results. We obtain a good recall of anomalies while having a very low false alarm rate (FPR) in a fully unsupervised procedure. We compare also with other anomaly detectors (NuPic, ADVec) from the state-of-the-art.
PPSN

Reinforcement learning with n-tuples on the game Connect-4

Markus Thill , Patrick Koch , and Wolfgang Konen

In PPSN’2012: 12th International Conference on Parallel Problem Solving From Nature, Taormina , 2012

Abs

Learning complex game functions is still a difficult task. We apply temporal difference learning (TDL), a well-known variant of the reinforcement learning approach, in combination with n-tuple networks to the game Connect-4. Our agent is trained just by self-play. It is able, for the first time, to consistently beat the optimal-playing Minimax agent (in game situations where a win is possible). The n-tuple network induces a mighty feature space: It is not necessary to design certain features, but the agent learns to select the right ones. We believe that the n-tuple network is an important ingredient for the overall success and identify several aspects that are relevant for achieving high-quality results. The architecture is sufficiently general to be applied to similar reinforcement learning tasks as well.