Search papers, labs, and topics across Lattice.
The paper introduces temporal-conditioned Normalizing Flows (tcNF), an autoregressive framework for time series anomaly detection that models temporal dependencies by conditioning flows on past observations. tcNF learns a probability distribution of expected behavior, enabling anomaly detection by flagging low-probability events. Experiments on diverse datasets demonstrate tcNF's accuracy and robustness compared to existing methods.
Time series anomaly detection gets a boost from temporal-conditioned normalizing flows that capture complex temporal dynamics and uncertainty.
This paper introduces temporal-conditioned normalizing flows (tcNF), a novel framework that addresses anomaly detection in time series data with accurate modeling of temporal dependencies and uncertainty. By conditioning normalizing flows on previous observations, tcNF effectively captures complex temporal dynamics and generates accurate probability distributions of expected behavior. This autoregressive approach enables robust anomaly detection by identifying low-probability events within the learned distribution. We evaluate tcNF on diverse datasets, demonstrating good accuracy and robustness compared to existing methods. A comprehensive analysis of strengths and limitations and open-source code is provided to facilitate reproducibility and future research.
