Research map / Time series / Causal and generative time series
Nonnegative Matrix Factorization for Time Series Recovery From a Few Temporal Aggregates
Jiali Mei, Yohann De Castro, Yannig Goude, Georges Hébrail
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add autocorrelation penalty (penalized HALS/NeNMF) to NMF from temporal aggregates (ablation) | HALS and NeNMF · Irish dataset, random and periodic observations | RRMSE | Improved |
| Use temporal aggregates as input via simplex projection in NMF instead of matrix entries | NMF for time series recovery · Random observation periods, all four datasets | RRMSE vs interpolation | Improved |
| Use nuclear-norm matrix completion (softImpute) on cumulative consumption matrix | matrix completion baseline · All four datasets | RRMSE | Hurt |
| Unpenalized NMF with periodic observations compared to interpolation (ablation) | HALS and NeNMF without penalty · Periodic observations, real datasets | RRMSE | No clear effect |
Closest papers by meaning
- Temporal Regularized Matrix Factorization for High-dimensional Time Series Prediction (NeurIPS 2016 · similarity 0.84)
- Regularization in Hierarchical Time Series Forecasting with Application to Electricity Smart Meter Data (AAAI 2017 · similarity 0.84)
- Matrix Factorisation for Scalable Energy Breakdown (AAAI 2017 · similarity 0.83)
- Non-Parametric Outliers Detection in Multiple Time Series A Case Study: Power Grid Data Analysis (AAAI 2018 · similarity 0.82)
- Deep Switching Auto-Regressive Factorization: Application to Time Series Forecasting (AAAI 2021 · similarity 0.82)
- PowerPM: Foundation Model for Power Systems (NeurIPS 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
