Research map / Time series / Causal and generative time series
GT-GAN: General Purpose Time Series Synthesis with Generative Adversarial Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 log-density MLE training of generator (ablation) | GT-GAN · Sines/Stocks/Energy/MuJoCo regular | discriminative score: .023 -> .012 (Sines) | Helped |
| add log-density MLE training (ablation) | GT-GAN · regular datasets | predictive score: .315 -> .312 (Energy) | No clear effect |
| add autoencoder pre-training (ablation) | GT-GAN · regular datasets | discriminative score: .046 -> .012 (Sines) | Helped |
| add autoencoder pre-training (ablation) | GT-GAN · regular datasets | predictive score: .299 -> .312 (Energy) | Mixed |
| add autoencoder to flow generator (vs w/o AE) (ablation) | GT-GAN · MuJoCo irregular 30% dropped | discriminative score: .500 -> .249 | Helped |
| add GAN+AE to flow-only (CTFP) model (ablation) | GT-GAN · MuJoCo irregular 30% dropped | discriminative score: .467 -> .249 | Helped |
| add flow/CTFP generator to AE-only model (ablation) | GT-GAN · MuJoCo irregular 30% dropped | discriminative score: .495 -> .249 | Helped |
| use GT-GAN instead of TimeGAN | GT-GAN · MuJoCo regular | discriminative score: .409 -> .245 | Helped |
| use GT-GAN instead of TimeGAN | GT-GAN · Sines regular | discriminative score: .011 -> .012 | No clear effect |
| use GT-GAN instead of GRU-D/delta-t baselines | GT-GAN · irregular 70% dropped | discriminative score: .436 -> .278 (Sines) | Helped |
Closest papers by meaning
- Time-series Generative Adversarial Networks (NeurIPS 2019 · similarity 0.89)
- Time Series Generation Under Data Scarcity: A Unified Generative Modeling Approach (NeurIPS 2025 · similarity 0.86)
- Generative Modeling of Regular and Irregular Time Series Data via Koopman VAEs (ICLR 2024 · similarity 0.86)
- Timer: Generative Pre-trained Transformers Are Large Time Series Models (ICML 2024 · similarity 0.85)
- TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting (ICLR 2024 · similarity 0.84)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (AAAI 2025 · similarity 0.84)
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.