方法对比
并排查看您选择的方法；存在差异的行会高亮显示。
| 动态变分推断× | 顺序蒙特卡洛× | |
|---|---|---|
| 领域 | 贝叶斯 | 贝叶斯 |
| 方法族 | Bayesian methods | Bayesian methods |
| 起源年份≠ | 2014–2015 | 1993 (particle filter); 2006 (SMC samplers) |
| 提出者≠ | Bayer, Osendorfer, Krishnan and colleagues | Gordon, Salmond & Smith (particle filter); Del Moral, Doucet & Jasra (SMC samplers) |
| 类型≠ | Bayesian approximate inference | Sequential Bayesian computation |
| 开创性文献≠ | Krishnan, R. G., Shalit, U., & Sontag, D. (2015). Deep Kalman Filters. NIPS 2015 Workshop on Advances in Approximate Bayesian Inference. link ↗ | Gordon, N. J., Salmond, D. J., & Smith, A. F. M. (1993). Novel approach to nonlinear/non-Gaussian Bayesian state estimation. IEE Proceedings F - Radar and Signal Processing, 140(2), 107–113. DOI ↗ |
| 别名 | sequential variational inference, temporal variational inference, variational inference for state-space models, DVI | SMC, particle filter, sequential importance resampling, SMC sampler |
| 相关 | 6 | 6 |
| 摘要≠ | Dynamic variational inference extends the variational inference framework to sequential and time-series settings by positing a structured approximate posterior that respects the temporal ordering of latent states. It jointly learns a generative model of how hidden states evolve over time and a recognition network that maps observed sequences back to those latent states, optimising a sequential evidence lower bound (ELBO). | Sequential Monte Carlo (SMC) is a family of simulation-based algorithms that approximate evolving probability distributions by propagating and reweighting a cloud of weighted random draws called particles. It handles nonlinear, non-Gaussian models and streams of data naturally, making it the method of choice for real-time state estimation and posterior approximation over complex distributions. |
| ScholarGate数据集 ↗ |
|
|