Research map / Probabilistic and scientific modeling / Variational and generative sampling
A General Method for Amortizing Variational Filtering
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 AVF with 2 inference iterations instead of the baseline filtering method | VRNN · TIMIT test | average filtering free energy per step | Helped |
| Use AVF with 5 inference iterations (ablation) | SRNN · JSB Chorales test | average filtering free energy per step | Helped |
| Increase the number of inference iterations during training (ablation) | VRNN with AVF · TIMIT validation | average free energy per step | Helped |
| Add successive inference iterations after the prior estimate (ablation) | VRNN with AVF · TIMIT validation | relative improvement in free energy | Helped |
| Use AVF instead of the originally proposed filtering procedure | SVG · KTH Actions | average filtering free energy per step | Helped |
| Use AVF instead of the originally proposed filtering procedure | VRNN and SRNN · TIMIT | average filtering free energy per step | No clear effect |
| Use AVF instead of the originally proposed filtering procedure | SRNN · polyphonic music data sets | average filtering free energy per step | No clear effect |
Closest papers by meaning
- Online Variational Filtering and Parameter Learning (NeurIPS 2021 · similarity 0.90)
- Amortized Inference Regularization (NeurIPS 2018 · similarity 0.90)
- Meta-Learning with Shared Amortized Variational Inference (ICML 2020 · similarity 0.90)
- Iterative Amortized Inference (ICML 2018 · similarity 0.90)
- Meta-Amortized Variational Inference and Learning (AAAI 2020 · similarity 0.89)
- Amortized Structured Stochastic Variational Inference for Gaussian Process Latent Variable Models (NeurIPS 2026 · similarity 0.88)
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.