Research map / Probabilistic and scientific modeling / Variational and generative sampling
Rescuing neural spike train models from bad MLE
Diego Arribas, Yuan Zhao, Ill Memming Park
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 |
|---|---|---|---|
| Add a model-based MMD penalty to the negative log-likelihood | GLM on Monkey PMv and human neocortex data · Monkey PMv; Human Cortex | firing-rate stability | Improved |
| Use MMD regularization instead of L2 regularization on history-filter coefficients (ablation) | GLM on Monkey PMv and human neocortex data · Monkey PMv; Human Cortex | likelihood | Improved |
| Add the MMD penalty using the Gaussian-smoothed spike-train feature map (Model 2) (ablation) | GLM on Monkey LIP · Monkey LIP | validation log-likelihood | Improved |
| Use the mean over time of the history term as the feature map (Model 3) (ablation) | MMD-GLM · Monkey LIP | runaway self-excitation probability: 4.4% -> 0.1% | Improved |
| Increase the MMD weight using the temporal autocorrelation of the GLM history term (Model 5 versus Model 4) (ablation) | MMD-GLM · Monkey LIP | runaway self-excitation probability | Improved |
| Increase the MMD weight using the temporal autocorrelation of the GLM history term (Model 5 versus Model 4) (ablation) | MMD-GLM · Monkey LIP | fit to other statistics | Hurt |
| Use different spike-train kernels to define the MMD objective (ablation) | MMD-GLM · Monkey LIP | fit across spike-train statistics | Mixed |
Closest papers by meaning
- Enabling hyperparameter optimization in sequential autoencoders for spiking neural data (NeurIPS 2019 · similarity 0.85)
- Real Spike: Learning Real-Valued Spikes for Spiking Neural Networks (ECCV 2022 · similarity 0.85)
- P-SPIKESSM: HARNESSING PROBABILISTIC SPIKING STATE SPACE MODELS FOR LONG-RANGE DEPENDENCY TASKS (ICLR 2025 · similarity 0.84)
- Spike Distance Function as a Learning Objective for Spike Prediction (ICML 2024 · similarity 0.84)
- Enhancing Training of Spiking Neural Network with Stochastic Latency (AAAI 2024 · similarity 0.83)
- Inferring stochastic low-rank recurrent neural networks from neural data (NeurIPS 2024 · 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.
