Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
167 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Regime Learning for Differentiable Particle Filters (2405.04865v3)

Published 8 May 2024 in cs.LG and eess.SP

Abstract: Differentiable particle filters are an emerging class of models that combine sequential Monte Carlo techniques with the flexibility of neural networks to perform state space inference. This paper concerns the case where the system may switch between a finite set of state-space models, i.e. regimes. No prior approaches effectively learn both the individual regimes and the switching process simultaneously. In this paper, we propose the neural network based regime learning differentiable particle filter (RLPF) to address this problem. We further design a training procedure for the RLPF and other related algorithms. We demonstrate competitive performance compared to the previous state-of-the-art algorithms on a pair of numerical experiments.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (24)
  1. N. Gordon, D. Salmond, and A. Smith, “Novel approach to nonlinear/non-gaussian bayesian state estimation,” IEEE Proc. F, Radar and Sig. Processing, vol. 140, no. 2, pp. 107–113, 1993.
  2. R. Jinan and T. Raveendran, “Particle filters for multiple target tracking,” Procedia Technology, vol. 24, pp. 980–987, 2016.
  3. R. Jonschkowski, D. Rastogi, and O. Brock, “Differentiable particle filters: End-to-end learning with algorithmic priors,” in Proc. Robot.: Sci. and Syst., Pittsburgh, PA, USA, July 2018.
  4. P. Karkus, D. Hsu, and W. S. Lee, “Particle filter networks with application to visual localization,” in Proc. Conf. on Robot Learning, Zurich, CH, Oct 2018, pp. 169–178.
  5. E. Vankov, M. Guindani, and K. Ensor, “Filtering and estimation for a class of stochastic volatility models with intractable likelihoods,” Bayesian Analysis, vol. 14, Mar 2018.
  6. X. Chen and Y. Li, “Normalising flow-based differentiable particle filters,” 2024, arXiv:2403.01499.
  7. S. Mcginnity and G. Irwin, “A multiple model extension to the bootstrap filter for manoeuvring targets,” in IFAC Proc. Workshop Alg. & Arch. for Real Time Contr. (AARTC), Cancun, Mexico, April 1998, pp. 23–28.
  8. Y. El-Laham, L. Yang, P. M. Djurić, and M. F. Bugallo, “Particle filtering under general regime switching,” in Proc. Euro. Sig. Process. Conf. (EUSIPCO), Amsterdam, NL, Jan 2021, pp. 2378–2382.
  9. W. Li, X. Chen, W. Wang, V. Elvira, and Y. Li, “Differentiable bootstrap particle filters for regime-switching models,” in Proc. IEEE Stat. Sig. Process. Workshop (SSP), Hanoi, Vietnam, 2023, pp. 200–204.
  10. L. Martino, J. Read, V. Elvira, and F. Louzada, “Cooperative parallel particle filters for online model selection and applications to urban mobility,” Digit. Sig. Process., vol. 60, pp. 172–185, 2017.
  11. N. Chopin and O. Papaspiliopoulos, “Importance sampling,” in An Introduction to Sequential Monte Carlo.   New York, USA: Springer International Publishing, 2020, pp. 81–103.
  12. P. Del Moral and A. Doucet, “Particle methods: An introduction with applications,” in ESAIM: Proc., vol. 44, Jan 2014.
  13. N. Chopin and O. Papaspiliopoulos, “Particle filtering,” in An Introduction to Sequential Monte Carlo.   New York, USA: Springer International Publishing, 2020, pp. 129–165.
  14. N. Kantas, A. Doucet, S. Singh, and J. Maciejowski, “An overview of sequential monte carlo methods for parameter estimation in general state-space models,” in IFAC Proc. Volumes, vol. 42, no. 10, Saint-Malo, FR, 2009, pp. 774–785.
  15. A. Corenflos, J. Thornton, G. Deligiannidis, and A. Doucet, “Differentiable particle filtering via entropy-regularized optimal transport,” in Proc. Int. Conf. on Machine Learn. (ICML), Online, 18–24 Jul 2021, pp. 2100–2111.
  16. X. Chen and Y. Li, “An overview of differentiable particle filters for data-adaptive sequential bayesian inference,” Founds. of Data Sci., Oct 2023.
  17. A. Mastrototaro and J. Olsson, “Online variational sequential monte carlo,” 2023, arXiv:2312.12616.
  18. A. Campbell, Y. Shi, T. Rainforth, and A. Doucet, “Online variational filtering and parameter learning,” in Proc. Adv. in Neural Info. Process. Syst. (NeurIPS), Online, 2021, pp. 18 633–18 645.
  19. M. Zhu, K. P. Murphy, and R. Jonschkowski, “Towards differentiable resampling,” 2020, arXiv:2004.11938.
  20. N. Chopin and O. Papaspiliopoulos, “Convergence and stability of particle filters,” in An Introduction to Sequential Monte Carlo.   New York, USA: Springer International Publishing, 2020, pp. 167–188.
  21. H. Sak, A. Senior, and F. Beaufays, “Long short-term memory based recurrent neural network architectures for large vocabulary speech recognition,” 2014, arXiv:1402.1128.
  22. J. Carpenter, P. Clifford, and P. Fearnhead, “Improved particle filter for nonlinear problems,” in IEE Proc. Radar, Sonar and Navi., vol. 146, February 1999, pp. 2–7(5).
  23. A. Vaswani et al., “Attention is all you need,” in Adv. in Neural Info. Process. Syst. (NeurIPS), Long Beach, CA, USA, 2017.
  24. I. Loshchilov and F. Hutter, “Decoupled weight decay regularization,” 2017, arXiv:1711.05101.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets