Markus Thill

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.

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.

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.