# Enhanced SMC2: Leveraging Gradient Information from Differentiable Particle Filters Within Langevin Proposals

arXiv: https://arxiv.org/abs/2407.17296 · HTML: https://mlsanity.com/paper/2407.17296/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 40 references that are in the map.
- First posted on arXiv 24 Jul 2024.

## Builds on (most cited first, 12 of 12)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Particle Filter Networks with Application to Visual Localization](https://mlsanity.com/paper/1805.08975/) (2018)
- [Differentiable Particle Filtering via Entropy-Regularized Optimal Transport](https://mlsanity.com/paper/2102.07850/) (2021)
- [Automatic Differentiation of Programs with Discrete Randomness](https://mlsanity.com/paper/2210.08572/) (2022)
- [An overview of differentiable particle filters for data-adaptive sequential Bayesian inference](https://mlsanity.com/paper/2302.09639/) (2023)
- [Differentiable Particle Filtering without Modifying the Forward Pass](https://mlsanity.com/paper/2106.10314/) (2021)
- [Efficient Learning of the Parameters of Non-Linear Models using Differentiable Resampling in Particle Filters](https://mlsanity.com/paper/2111.01409/) (2021)
- [Getting Started with Particle Metropolis-Hastings for Inference in Nonlinear Dynamical Models](https://mlsanity.com/paper/1511.01707/) (2015)
- [Particle Metropolis-Hastings using gradient and Hessian information](https://mlsanity.com/paper/1311.0686/) (2013)
- [Particle-Based Score Estimation for State Space Model Learning in Autonomous Driving](https://mlsanity.com/paper/2212.06968/) (2022)
- [Particle Metropolis adjusted Langevin algorithms for state space models](https://mlsanity.com/paper/1402.0694/) (2014)
- [Revisiting semi-supervised training objectives for differentiable particle filters](https://mlsanity.com/paper/2405.01251/) (2024)

## Built on by (most cited first, 4 of 4)

- [Incorporating the ChEES Criterion into Sequential Monte Carlo Samplers](https://mlsanity.com/paper/2504.02627/) (2025)
- [Hess-MC2: Sequential Monte Carlo Squared using Hessian Information and Second Order Proposals](https://mlsanity.com/paper/2507.07461/) (2025)
- [Utilising Gradient-Based Proposals Within Sequential Monte Carlo Samplers for Training of Partial Bayesian Neural Networks](https://mlsanity.com/paper/2505.03797/) (2025)
- [Investigating Batch Inference in a Sequential Monte Carlo Framework for Neural Networks](https://mlsanity.com/paper/2601.21983/) (2026)

## Abstract (arXiv)

Sequential Monte Carlo Squared (SMC2) is a Bayesian method which can infer the states and parameters of non-linear, non-Gaussian state-space models. The standard random-walk proposal in SMC2 faces challenges, particularly with high-dimensional parameter spaces. This study outlines a novel approach by harnessing first-order gradients derived from a Common Random Numbers - Particle Filter (CRN-PF) using PyTorch. The resulting gradients can be leveraged within a Langevin proposal without accept/reject. Including Langevin dynamics within the proposal can result in a higher effective sample size and more accurate parameter estimates when compared with the random-walk. The resulting algorithm is parallelized on distributed memory using Message Passing Interface (MPI) and runs in O(log2N) time complexity. Utilizing 64 computational cores we obtain a 51x speed-up when compared to a single core. A GitHub link is given which provides access to the code.