mlsanityData to 7 Oct 2026

Differentiable Particle Filtering without Modifying the Forward Pass

arXiv 2106.10314 ↗First posted 18 Jun 2021stat.ML cs.LGcited by 18

Not on the map. Why?

Abstract (arXiv)

Particle filters are not compatible with automatic differentiation due to the presence of discrete resampling steps. While known estimators for the score function, based on Fisher's identity, can be computed using particle filters, up to this point they required manual implementation. In this paper we show that such estimators can be computed using automatic differentiation, after introducing a simple correction to the particle weights. This correction utilizes the stop-gradient operator and does not modify the particle filter operation on the forward pass, while also being cheap and easy to compute. Surprisingly, with the same correction automatic differentiation also produces good estimators for gradients of expectations under the posterior. We can therefore regard our method as a general recipe for making particle filters differentiable. We additionally show that it produces desired estimators for second-order derivatives and how to extend it to further reduce variance at the expense of additional computation.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 23 of its 51 references that are in the map. First posted on arXiv 18 Jun 2021.

Builds on

23 of 51 references.

References · 23

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,135
  3. 2015Automatic differentiation in machine learning: a surveycited by 846
  4. 2015Importance Weighted Autoencoderscited by 454
  5. 2013Black Box Variational Inferencecited by 317
  6. 2019INTERACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Mapscited by 231
  7. 2018Deep Variational Reinforcement Learning for POMDPscited by 122
  8. 2015Gradient Estimation Using Stochastic Computation Graphscited by 120
  9. 2017Variational Sequential Monte Carlocited by 93
  10. 2017Filtering Variational Objectivescited by 91
  11. 2018Tighter Variational Bounds are Not Necessarily Bettercited by 76
  12. 2017Auto-Encoding Sequential Monte Carlocited by 64
11 more
  1. 2018Particle Filter Networks with Application to Visual Localizationcited by 64
  2. 2021Differentiable Particle Filtering via Entropy-Regularized Optimal Transportcited by 59
  3. 2018Differentiable Particle Filters: End-to-End Learning with Algorithmic Priorscited by 58
  4. 2018DiCE: The Infinitely Differentiable Monte-Carlo Estimatorcited by 46
  5. 2018Pathwise Derivatives Beyond the Reparameterization Trickcited by 36
  6. 2015Nested Sequential Monte Carlo Methodscited by 19
  7. 2020Towards Differentiable Resamplingcited by 19
  8. 2016Stochastic Backpropagation through Mixture Density Distributionscited by 14
  9. 2019Loaded DiCE: Trading off Bias and Variance in Any-Order Score Function Estimators for Reinforcement Learningcited by 12
  10. 2013Parallel resampling in the particle filtercited by 3
  11. 2018Revisiting Reweighted Wake-Sleep for Models with Stochastic Control Flowcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 2; 2023: 0; 2024: 8; 2025: 5; 2026: 2.

18 papers.

Sort:

By area:

  1. 2022Automatic Differentiation of Programs with Discrete Randomnesscited by 28
  2. 2024Normalizing Flow-based Differentiable Particle Filterscited by 10
  3. 2021Efficient Learning of the Parameters of Non-Linear Models using Differentiable Resampling in Particle Filterscited by 9
  4. 2024Sequential Monte Carlo for Inclusive KL Minimization in Amortized Variational Inferencecited by 7
  5. 2024Enhanced SMC2: Leveraging Gradient Information from Differentiable Particle Filters Within Langevin Proposalscited by 4
  6. 2025Learning to be Smooth: An End-to-End Differentiable Particle Smoothercited by 4
  7. 2022Particle-Based Score Estimation for State Space Model Learning in Autonomous Drivingcited by 3
  8. 2024Differentiable and Stable Long-Range Tracking of Multiple Posterior Modescited by 2
  9. 2025Deep Variational Sequential Monte Carlo for High-Dimensional Observationscited by 2
  10. 2025DiffPF: Differentiable Particle Filtering with Generative Sampling via Conditional Diffusion Modelscited by 2
  11. 2025PyDPF: A Python Package for Differentiable Particle Filteringcited by 2
  12. 2025Diffusion differentiable resamplingcited by 2
6 more
  1. 2026Auto-differentiable data assimilation: Co-learning of states, dynamics, and filtering algorithmscited by 1
  2. 2024Differentiable Particle Filtering using Optimal Placement Resamplingcited by 0
  3. 2024Accelerated Inference for Partially Observed Markov Processes using Automatic Differentiationcited by 0
  4. 2024Learning Flock: Enhancing Sets of Particles for Multi~Sub-State Particle Filtering with Neural Augmentationcited by 0
  5. 2024Learning state and proposal dynamics in state-space models using differentiable particle filters and neural networkscited by 0
  6. 2026Efficient Learning of Deep State Space Models via Importance Smoothingcited by 0