Research map / Probabilistic and scientific modeling / Dynamical systems
InputDSA: Demixing, then comparing recurrent and externally driven dynamics
Ann Huang, Mitchell Ostrow, Satpreet Harcharan Singh, Leo Kozachkov, Ila R Fiete, Kanaka Rajan
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 |
|---|---|---|---|
| replace DMDc with SubspaceDMDc for state distance under partial observation (ablation) | InputDSA · Simulated partially observed systems, state distance | silhouette score: 0.68 -> 0.94 | Improved |
| replace DMD with SubspaceDMDc for state distance under partial observation (ablation) | InputDSA · Simulated partially observed systems, state distance | silhouette score: 0.6 -> 0.94 | Improved |
| replace DMDc with SubspaceDMDc for input distance under partial observation (ablation) | InputDSA · Simulated partially observed systems, input distance | silhouette score: 0.19 -> 0.83 | Improved |
| corrupt the input with noise or polynomial transforms, down to SER near 1 (ablation) | InputDSA SubspaceDMDc · Input noise robustness | correlation with true-input distances: correlations remain above r > 0.75 as SER approaches 1 | No clear effect |
| use surrogate or random inputs instead of ground-truth input (ablation) | InputDSA state distance · Random Target Reach RNNs | correlation with ground-truth distances | No clear effect |
| use surrogate inputs that correlate more with the true input (e.g. action plus position) (ablation) | InputDSA input distance · Random Target Reach RNNs | correlation with ground-truth distances | Improved |
| compare Top vs Bottom plume agents on input-driven dynamics rather than intrinsic dynamics | InputDSA · Plume Tracking | group separation of distances | Improved |
| increase system size from 2 to 1000 dimensions with 2 observed (ablation) | SubspaceDMDc state distance · Simulated systems | silhouette score | Hurt |
Closest papers by meaning
- KoopSTD: Reliable Similarity Analysis between Dynamical Systems via Approximating Koopman Spectrum with Timescale Decoupling (ICML 2025 · similarity 0.87)
- Inference of Neural Dynamics Using Switching Recurrent Neural Networks (NeurIPS 2024 · similarity 0.87)
- Beyond DSA: Conjugacy-based Comparison of Dynamical Systems (NeurIPS 2026 · similarity 0.87)
- Learning Interpretable Hierarchical Dynamical Systems Models from Time Series Data (ICLR 2025 · similarity 0.86)
- Meta-Dynamical State Space Models for Integrative Neural Data Analysis (ICLR 2025 · similarity 0.85)
- Out-of-Domain Generalization in Dynamical Systems Reconstruction (ICML 2024 · similarity 0.85)
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.
