Research map / Time series / Causal and generative time series
DeNOTS: Stable Deep Neural ODEs for Time Series
Ilya Kuleshov, Evgenia Romanenkova, Vladislav Andreevich Zhuzhel, Galina Boeva, Evgeni Vorsin, Alexey Zaytsev
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 |
|---|---|---|---|
| scale integration time (SNCDE) with Anti-NF (DeNOTS) vs other baselines | Neural CDE · UWGL, InsectSound, Pendulum, Sepsis | mean rank: best baseline mean rank 4.0 -> 1.0 | Improved |
| add negative feedback (Anti-NF or Sync-NF) to scaled neural CDE (ablation) | SNCDE · Pendulum drop attack | R^2: 0.40 -> 0.61 (Anti-NF), 0.64 (Sync-NF) | Improved |
| add negative feedback under Gaussian noise attack (ablation) | SNCDE · Pendulum change attack | R^2: 0.51 -> 0.64 (Anti-NF), 0.61 (Sync-NF) | Improved |
| use strict Sync-NF instead of Anti-NF (ablation) | SNCDE · SineMix long-term memory | R^2: 1 -> 0.3 | Hurt |
| increase NFE by scaling time instead of lowering tolerance (with Anti-NF) (ablation) | SNCDE · Pendulum | Pearson corr log-NFE vs R^2: 0.5 -> 0.9 | Improved |
| increase NFE by scaling time (without NF) (ablation) | SNCDE No NF · Pendulum | Pearson corr log-NFE vs R^2: 0.4 -> -0.5 | Hurt |
| Anti-NF instead of Sync-NF (ablation) | SNCDE · Pendulum | R^2 | No clear effect |
| larger time scale D lowers weight norms (ablation) | SNCDE tanh · Pendulum | l2 norm of weights | Improved |
Closest papers by meaning
- Stable Neural Stochastic Differential Equations in Analyzing Irregular Time Series Data (ICLR 2024 · similarity 0.89)
- On the Generalization and Approximation Capacities of Neural Controlled Differential Equations (ICLR 2024 · similarity 0.88)
- Neural Stochastic Flows: Solver-Free Modelling and Inference for SDE Solutions (NeurIPS 2025 · similarity 0.87)
- Global Convergence in Neural ODEs: Impact of Activation Functions (ICLR 2025 · similarity 0.86)
- How Deep Do We Need: Accelerating Training and Inference of Neural ODEs via Control Perspective (ICML 2024 · similarity 0.86)
- Space-Time Continuous PDE Forecasting using Equivariant Neural Fields (NeurIPS 2024 · similarity 0.86)
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.
