"In this paper, the forgetting of the initial distribution for a non-ergodic Hidden Markov Models (HMM) is studied. A new set of conditions is proposed to establish the forgetting property of the filter, which significantly extends all the existing results. Both a pathwise-type convergence of the total variation distance of the filter started from two different initial distributions, and a convergence in expectation are considered. The results are illustrated using generic models of non-ergodic HMM and extend all the results known so far."

"We present an online version of the expectation-maximization (EM) algorithm for hidden Markov models (HMMs). The sufficient statistics required for parameters estimation is computed recursively with time, that is, in an online way instead of using the batch forward-backward procedure. This computational scheme is generalized to the case where the model parameters can change with time by introducing a discount factor into the recurrence relations. The resulting algorithm is equivalent to the batch EM algorithm, for appropriate discount factor and scheduling of parameters update. On the other hand, the online algorithm is able to deal with dynamic environments, i.e., when the statistics of the observed data is changing with time. The implications of the online algorithm for probabilistic modeling in neuroscience are briefly discussed."

"Computing smoothing distributions, the distributions of one or more states conditional on past, present, and future observations is a recurring problem when operating on general hidden Markov models. The aim of this paper is to provide a foundation of particle-based approximation of such distributions and to analyze, in a common unifying framework, different schemes producing such approximations. In this setting, general convergence results, including exponential deviation inequalities and central limit theorems, are established. In particular, time uniform bounds on the marginal smoothing error are obtained under appropriate mixing conditions on the transition kernel of the latent chain. In addition, we propose an algorithm approximating the joint smoothing distribution at a cost that grows only linearly with the number of particles."

"We study online learning of linear and kernel-based predictors, when individual examples are corrupted by random noise, and both examples and noise type can be chosen adversarially and change over time. We begin with the setting where some auxiliary information on the noise distribution is provided, and we wish to learn predictors with respect to the squared loss. Depending on the auxiliary information, we show how one can learn linear and kernel-based predictors, using just 1 or 2 noisy copies of each example. We then turn to discuss a general setting where virtually nothing is known about the noise distribution, and one wishes to learn with respect to general losses and using linear and kernel-based predictors. We show how this can be achieved using a random, essentially constant number of noisy copies of each example. Allowing multiple copies cannot be avoided: Indeed, we show that the setting becomes impossible when only one noisy copy of each instance can be accessed. To obtain our results we introduce several novel techniques, some of which might be of independent interest."

"In this paper we introduce a projection method for the space of probability distributions based on the differential geometric approach to statistics. This method is based on a direct L2 metric as opposed to the usual Hellinger distance and the related Fisher Information metric. We explain how this apparatus can be used for the nonlinear filtering problem, in relationship also to earlier projection methods based on the Fisher metric. Past projection filters focused on the Fisher metric and the exponential families that made the filter correction step exact. In this work we introduce the mixture projection filter, namely the projection filter based on the direct $L^2$ metric and based on a manifold given by a mixture of pre-assigned densities. The resulting prediction step in the filtering problem is described by a linear differential equation, while the correction step can be made exact."

"In this paper we study ergodic properties of hidden Markov models with a generalized observation structure. In particular sufficient conditions for the existence of a unique invariant measure for the pair filter-observation are given. Furthermore, necessary and sufficient conditions for the existence of a unique invariant measure of the triple state-observation-filter are provided in terms of asymptotic stability in probability of incorrectly initialized filters. We also study the asymptotic properties of the filter and of the state estimator based on the observations as well as on the knowledge of the initial state. Their connection with minimal and maximal invariant measures is also studied."

"The Expectation Maximization (EM) algorithm is a versatile tool for model parameter estimation in latent data models. When processing large data sets or data stream however, EM becomes intractable since it requires the whole data set to be available at each iteration of the algorithm. In this contribution, a new generic online EM algorithm for model parameter inference in general Hidden Markov Model is proposed. This new algorithm updates the parameter estimate after a block of observations is processed (online). The convergence of this new algorithm is established, and the rate of convergence is studied showing the impact of the block size. An averaging procedure is also proposed to improve the rate of convergence. Finally, practical illustrations are presented as well as extensions to some online stochastic EM when Sequential Monte Carlo methods have to be used in combination, in order to make the E-step tractable."

"Particle learning (PL) provides state filtering, sequential parameter learning and smoothing in a general class of state space models. Our approach extends existing particle methods by incorporating the estimation of static parameters via a fully-adapted filter that utilizes conditional sufficient statistics for parameters and/or states as particles. State smoothing in the presence of parameter uncertainty is also solved as a by-product of PL. In a number of examples, we show that PL outperforms existing particle filtering alternatives and proves to be a competitor to MCMC."

Huzzah!

I think you mis-spelled "smoothing spline". HTH. HAND.

"Sequential Bayesian filters, such as particle filters, are often presented as an ideal means of tracking the state of nonlinear systems. Here shadowing filters are demonstrated to perform better than sequential filters at tracking under specific circumstances. The success of shadowing filters is attributed to avoiding both well-known deficiencies of particle filters, and some newly identified problems." Huh.

So, how _do_ you aim for the stars and/or make sure you hit London?

State-space modeling with linear/Gaussian state evolution and generalized linear models for the observations. Looks reasonable, lacks a few improvements like diffuse initial conditions in the Kalman filter.

Andy's book is appearing in print at last. (SIAM seems to indicate it's available now, the usual online bookstores say not until the end of the year.)

When can the state of one process be recovered without error from another? (Use of infinite time limit here is not quite relevant to my immediate needs so must see how to modify proof.)

"This paper develops a connection between the asymptotic stability of nonlinear filters and a notion of observability. We consider a general class of hidden Markov models in continuous time with compact signal state space, and call such a model observable if no two initial measures of the signal process give rise to the same law of the observation process. We demonstrate that observability implies stability of the filter, i.e., the filtered estimates become insensitive to the initial measure at large times. For the special case where the signal is a finite-state Markov process and the observations are of the white noise type, a complete (necessary and sufficient) characterization of filter stability is obtained in terms of a slightly weaker detectability condition. In addition to observability, the role of controllability in filter stability is explored. Finally, the results are partially extended to non-compact signal state spaces."