Get Started
Home
Topics
Search
Library
Research questionHow can synthetic time-series generation preserve temporal and frequency structure when training data are scarce?Small datasets make it difficult for generative models to reproduce both long- and short-range behavior. Modeling only one representation can also lose structure expressed in the complementary temporal or frequency domain.
AI
Diffusion Models
Machine Learning
Statistical Machine Learning
Latest papersRecent research connected to this question, newest first.WaveletDiff: Multilevel Wavelet Diffusion For Time Series GenerationThe source trains diffusion models on wavelet coefficients and evaluates them on real-world datasets from energy, finance, and neuroscience. The evidence is limited to those datasets and generation evaluations; broader deployment value and downstream-task effects are not established.research paper · Sep 4, 2026
Related questions
How can synthetic medical time-series generation preserve rare-class patterns across temporal scales for downstream prediction?How can multimodal time-series forecasting align sparse text with heterogeneous temporal dynamics?How can lightweight time-series forecasters learn accurate, specialized forecasts from scarce private data?How can multivariate time-series imputers recover structured gaps under high missingness?