Research map / Probabilistic and scientific modeling / Dynamical systems
Dynamic Ensemble Modeling Approach to Nonstationary Neural Decoding in Brain-Computer Interfaces
Yu Qi, Bin Liu, Yueming Wang, Gang Pan
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 |
|---|---|---|---|
| Use dynamic Bayesian model selection and combination instead of a fixed measurement model | DyEnsemble · Rat 1 and Rat 2 neural decoding, noisy neurons | CC: Rat 1 Noisy (#4): 0.671 vs 0.560; Rat 2 Noisy (#4): 0.596 vs 0.381 | Improved |
| Add weight perturbation (p=0.1) (ablation) | DyEnsemble · Rat 2, Noisy (#2) | CC: 0.579$\pm$0.066 -> 0.665$\pm$0.024 | Improved |
| Add weight perturbation (p=0.1) (ablation) | DyEnsemble · Rat 1, Noisy (#4) | CC: 0.558$\pm$0.009 -> 0.557$\pm$0.035 | Hurt |
| Add neuron dropout of 5 neurons with perturbation (ablation) | DyEnsemble · Rat 1, Noisy (#4) | CC: 0.557$\pm$0.035 -> 0.671$\pm$0.039 | Improved |
| Add neuron dropout of 5 neurons with perturbation (ablation) | DyEnsemble · Rat 2, Original | CC: 0.780$\pm$0.006 -> 0.803$\pm$0.009 | Mixed |
| Drop 5 rather than 2 neurons in candidate models (ablation) | DyEnsemble · Rat 2, Noisy (#4) | CC: 0.516$\pm$0.092 -> 0.596$\pm$0.035 | Improved |
| Increase candidate model size (ablation) | DyEnsemble · Rat 1 and Rat 2 neural decoding | CC | Mixed |
| Increase candidate model count from 5 to 20 (ablation) | DyEnsemble · Rat 1 and Rat 2 neural decoding | CC | Improved |
| Increase candidate model count above 20 (ablation) | DyEnsemble · Rat 1 and Rat 2 neural decoding | CC | No clear effect |
| Use a smaller forgetting factor (ablation) | DyEnsemble · Rat 1 and Rat 2 neural decoding | CC | Improved |
| Increase weight perturbation from 0.01 to 0.1 (ablation) | DyEnsemble · Rat 1 and Rat 2 neural decoding | CC | Improved |
| Increase weight perturbation above 0.2 (ablation) | DyEnsemble · Rat 1 and Rat 2 neural decoding | CC | Hurt |
Closest papers by meaning
- Exploring the trade-off between deep-learning and explainable models for brain-machine interfaces (NeurIPS 2024 · similarity 0.86)
- SPINT: Spatial Permutation-Invariant Neural Transformer for Consistent Intracortical Motor Decoding (NeurIPS 2025 · similarity 0.84)
- A Scalable, Causal, and Energy Efficient Framework for Neural Decoding with Spiking Neural Networks (NeurIPS 2025 · similarity 0.84)
- Few-shot Algorithms for Consistent Neural Decoding (FALCON) Benchmark (NeurIPS 2024 · similarity 0.83)
- CREIMBO: Cross-Regional Ensemble Interactions in Multi-view Brain Observations (ICLR 2025 · similarity 0.83)
- Flow Matching for Few-Trial Neural Adaptation with Stable Latent Dynamics (ICML 2025 · similarity 0.83)
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.
