Online false discovery rate control for anomaly detection in time series

Authors: Quentin Rebjock, Baris Kurt, Tim Januschowski, Laurent Callot

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We show the soundness of these rules in both theory and experiments. We evaluate experimentally the performances of the proposed algorithms and demonstrate that we can overcome the challenges occurring when alternative hypotheses are exceedingly rare.
Researcher Affiliation Collaboration Quentin Rebjock EPFL; Barıs Kurt Amazon Research; Tim Januschowski Amazon Research; Laurent Callot Amazon Research
Pseudocode No The paper describes algorithms and derivations but does not include any explicitly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any concrete access information (e.g., repository links, explicit statements of code release) for the methodology's source code.
Open Datasets Yes For these experiments we generate data following a mixture of distribution... We next demonstrate the performance of our methods on p-values generated by a simple forecaster applied on the Server Machine Dataset (SMD) [28].
Dataset Splits No The paper describes generating synthetic data and using the SMD dataset, but it does not specify explicit training, validation, or test splits (e.g., percentages or counts) for these datasets.
Hardware Specification No The paper does not provide specific details about the hardware (e.g., CPU, GPU models, memory) used to run the experiments.
Software Dependencies No The paper does not specify any software dependencies with version numbers (e.g., Python, PyTorch, or other libraries).
Experiment Setup Yes For all models, the target FDR is 0.1; for SAFFRON, λ = 1/2, and for ADDIS λ = 1/4. We set δ to 0.99 for Decay Lord and Decay Saffron models.