mlsanityData to 7 Oct 2026

TimeInf: Time Series Data Contribution via Influence Functions

arXiv 2407.15247 ↗First posted 21 Jul 2024cs.LG stat.MLcited by 6

Not on the map. Why?

Abstract (arXiv)

Evaluating the contribution of individual data points to a model's prediction is critical for interpreting model predictions and improving model performance. Existing data contribution methods have been applied to various data types, including tabular data, images, and text; however, their primary focus has been on i.i.d. settings. Despite the pressing need for principled approaches tailored to time series datasets, the problem of estimating data contribution in such settings remains under-explored, possibly due to challenges associated with handling inherent temporal dependencies. This paper introduces TimeInf, a model-agnostic data contribution estimation method for time-series datasets. By leveraging influence scores, TimeInf attributes model predictions to individual time points while preserving temporal structures between the time points. Our empirical results show that TimeInf effectively detects time series anomalies and outperforms existing data attribution techniques as well as state-of-the-art anomaly detection methods. Moreover, TimeInf offers interpretable attributions of data values, allowing us to distinguish diverse anomalous patterns through visualizations. We also showcase a potential application of TimeInf in identifying mislabeled anomalies in the ground truth annotations.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 27 of its 57 references that are in the map. First posted on arXiv 21 Jul 2024.

Builds on

27 of 57 references.

References · 27

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  2. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  3. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  4. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,400
  5. 2017Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networkscited by 869
  6. 2019Data Shapley: Equitable Valuation of Data for Machine Learningcited by 617
  7. 2020Estimating Training Data Influence by Tracing Gradient Descentcited by 565
  8. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 492
  9. 2018Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholdingcited by 407
  10. 2021Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancycited by 387
  11. 2023Studying Large Language Model Generalization with Influence Functionscited by 311
  12. 2022TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Datacited by 211
15 more
  1. 2020Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progresscited by 148
  2. 2019Efficient Task-Specific Data Valuation for Nearest Neighbor Algorithmscited by 147
  3. 2019On the Accuracy of Influence Functions for Measuring Group Effectscited by 146
  4. 2023DataInf: Efficiently Estimating Data Influence in LoRA-tuned LLMs and Diffusion Modelscited by 121
  5. 2023DCdetector: Dual Attention Contrastive Representation Learning for Time Series Anomaly Detectioncited by 103
  6. 2021Towards a Rigorous Evaluation of Time-series Anomaly Detectioncited by 95
  7. 2020TimeSHAP: Explaining Recurrent Models through Sequence Perturbationscited by 73
  8. 2023OpenDataVal: a Unified Benchmark for Data Valuationcited by 55
  9. 2024SoftPatch: Unsupervised Anomaly Detection with Noisy Datacited by 55
  10. 2024An Analysis of Linear Time Series Forecasting Modelscited by 40
  11. 2023Data-OOB: Out-of-bag Estimate as a Simple and Efficient Data Valuecited by 38
  12. 2024TimeSeriesBench: An Industrial-Grade Benchmark for Time Series Anomaly Detection Modelscited by 22
  13. 2020Approximate Cross-Validation for Structured Modelscited by 9
  14. 2022Feature Importance for Time Series Data: Improving KernelSHAPcited by 5
  15. 2014Most Convex Functions Have Unique Minimizerscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 2; 2026: 3.

6 papers.

  1. 2024Capturing the Temporal Dependence of Training Data Influencecited by 34
  2. 2025Rating Quality of Diverse Time Series Data by Meta-learning from LLM Judgmentcited by 1
  3. 2026Directional Influence Function: Estimating Training Data Influence in Constrained Learningcited by 1
  4. 2025Lightweight Time Series Data Valuation on Time Series Foundation Models via In-Context Finetuningcited by 0
  5. 2026TimeLAVA: Learning-Agnostic Valuation for Time Series Datacited by 0
  6. 2026Loss-Guided Pretraining Data Selection for Time-Series Foundation Modelscited by 0