Efficient Learning of the Parameters of Non-Linear Models using Differentiable Resampling in Particle Filters
arXiv:2111.01409 · doi:10.1109/TSP.2022.3187868
Abstract
It has been widely documented that the sampling and resampling steps in particle filters cannot be differentiated. The {\itshape reparameterisation trick} was introduced to allow the sampling step to be reformulated into a differentiable function. We extend the {\itshape reparameterisation trick} to include the stochastic input to resampling therefore limiting the discontinuities in the gradient calculation after this step. Knowing the gradients of the prior and likelihood allows us to run particle Markov Chain Monte Carlo (p-MCMC) and use the No-U-Turn Sampler (NUTS) as the proposal when estimating parameters. We compare the Metropolis-adjusted Langevin algorithm (MALA), Hamiltonian Monte Carlo with different number of steps and NUTS. We consider two state-space models and show that NUTS improves the mixing of the Markov chain and can produce more accurate results in less computational time.
arXiv comment35 pages, 10 figures
References in corpus (12)
- Auto-Encoding Variational Bayes
- Categorical Reparameterization with Gumbel-Softmax
- General state space Markov chains and MCMC algorithms
- A Conceptual Introduction to Hamiltonian Monte Carlo
- Randomized Hamiltonian Monte Carlo
- Geometric integrators and the Hamiltonian Monte Carlo method
- Particle Metropolis-Hastings using gradient and Hessian information
- Identifying the Optimal Integration Time in Hamiltonian Monte Carlo
- Towards Differentiable Resampling
- Model error covariance estimation in particle and ensemble Kalman filters using an online expectation-maximization algorithm
- Particle Metropolis adjusted Langevin algorithms for state space models
- Differentiable Particle Filtering without Modifying the Forward Pass