Research map / Probabilistic and scientific modeling / Variational and generative sampling
Collapsed variational Bayes for Markov jump processes
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 |
|---|---|---|---|
| collapsed variational inference with optimized time discretization instead of MCMC | Markov jump process posterior inference · Check-in data (FourSquare) | reconstruction error | Helped |
| random interval splitting instead of even splitting in VB (ablation) | VB time-discretization update · Check-in data (FourSquare) | reconstruction error | Helped |
| even interval splitting instead of random splitting (ablation) | VB time-discretization update · Check-in data (FourSquare) | running time | Helped |
| generalized uniformization with per-state Omega in MCMC (ablation) | MCMC baseline · Check-in data (FourSquare) | reconstruction error | No clear effect |
| fixing Omega to true value in VB (oracle) vs. estimated Omega (ablation) | VB on synthetic datasets · Synthetic dataset 1 and 2 | reconstruction error | No clear effect |
| VB produces fewer, non-redundant transitions vs MCMC | Markov jump process trajectory posterior · Check-in data (FourSquare) | number of transitions per trajectory | Helped |
| VB error gap to MCMC closes with more iterations and resampled data | VB vs MCMC · Check-in data (FourSquare) | reconstruction error | Mixed |
Closest papers by meaning
- Moment-Based Variational Inference for Markov Jump Processes (ICML 2019 · similarity 0.91)
- Variational Inference for Continuous-Time Switching Dynamical Systems (NeurIPS 2021 · similarity 0.90)
- Neural Markov Jump Processes (ICML 2023 · similarity 0.90)
- Foundation Inference Models for Markov Jump Processes (NeurIPS 2024 · similarity 0.88)
- Online Variational Sequential Monte Carlo (ICML 2024 · similarity 0.87)
- Piecewise deterministic generative models (NeurIPS 2024 · similarity 0.86)
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.