Research map / Probabilistic and scientific modeling / Variational and generative sampling
Structured Recognition for Generative Models with Explaining Away
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 |
|---|---|---|---|
| Use structured recognition that captures posterior correlations between latent processes | SR-nlGPFA vs SGP-VAE · Synthetic data | SMSE: 0.36± 0.05 -> 0.31±0.02 | Helped |
| Use structured recognition that captures posterior correlations between latent processes | SR-nlGPFA vs SGP-VAE · Synthetic data | NLL: 1.76± 0.07 -> 1.69± 0.04 | Helped |
| Use structured recognition that captures posterior correlations between latent processes | SR-nlGPFA vs SGP-VAE · EEG | SMSE: 0.35± 0.05 -> 0.27 ± 0.05 | Helped |
| Use structured recognition that captures posterior correlations between latent processes | SR-nlGPFA vs SGP-VAE · EEG | NLL: 2.17± 0.17 -> 1.81±0.21 | Helped |
| Use structured recognition that captures posterior correlations between latent processes | SR-nlGPFA vs SGP-VAE · Population spiking | SMSE: 0.55 ± 0.19 -> 0.47±0.18 | Helped |
| Use structured recognition rather than fully factorised recognition | SRVAE-GMM vs SIN-GMM · Pinwheel | variational free energy | Helped |
| Use structured recognition rather than fully factorised recognition | SRVAE-GMM vs SIN-GMM · Pinwheel | reconstruction MSE | Helped |
| Use structured recognition rather than fully factorised recognition | SRVAE-GMM vs SIN-GMM · Pinwheel | generation fidelity | Helped |
| Use SR-nlGPFA rather than SGP-VAE for latent trajectory recovery | Population spiking · Hippocampal CA1 and mEC recordings | direction modulation correlation: p-value=$5.56 \times 10^{-4}$ | Helped |
| Use structured rather than factorised recognition | SR-nlGPFA vs SGP-VAE · Population spiking | latent trajectory smoothness | Helped |
Closest papers by meaning
- Multi-modal Gaussian Process Variational Autoencoders for Neural and Behavioral Data (ICLR 2024 · similarity 0.88)
- Learning Structured Latent Factors from Dependent Data:A Generative Model Framework from Information-Theoretic Perspective (ICML 2020 · similarity 0.88)
- Composing graphical models with neural networks for structured representations and fast inference (NeurIPS 2016 · similarity 0.87)
- Learning to Discover Sparse Graphical Models (ICML 2017 · similarity 0.87)
- eXponential FAmily Dynamical Systems (XFADS): Large-scale nonlinear Gaussian state-space modeling (NeurIPS 2024 · similarity 0.86)
- CASL-VAE: Learning Structured Latent Variables from Unpaired Data for Semi-supervised Clustering and Paired Sample Generation (NeurIPS 2026 · 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.