betaFeedback

WorksMathematics

The Statistical Geometry of Latent Recovery in Hidden Markov Models

Abstract

Hidden Markov models (HMMs) offer an important capability to interpret fitted latent states and transitions. When emission distributions nearly coincide, a small observation-density error does not ensure accurate estimation of the latent quantities needed for interpretation. This motivates us to characterize convergence rates for latent recovery that account for emission separation. Our main contribution lies in deriving deterministic inverse inequalities that convert Hellinger error between fitted and true densities into bounds on latent recovery errors. We develop these bounds for multivariate Gaussian HMMs with a common, known emission covariance matrix and combine them with a Hellinger estimation guarantee for the global maximum likelihood estimator to obtain latent recovery rates. Although the joint observation distribution is a Gaussian mixture with exponentially many path components, they share the same emission centers across time. Leveraging this shared structure, we show that the separation exponents are governed by the maximum number of simultaneously approaching centers rather than the number of path components. Under exact specification, the global bounds on state-sequence probability discrepancies and the 1-Wasserstein loss for latent location distributions exhibit sharp separation exponents that do not increase with the number of selected times. Moreover, sufficiently small observation-density error guarantees unique matching between fitted and true emission centers, which yields error bounds for individual emission centers and transition rows. Under over-specification, individual split-state transition rows are not necessarily identifiable, so we quantify aggregate latent recovery errors and show that state-sequence probability bounds incur larger separation exponents compared to exact specification. We conduct synthetic experiments to validate our theoretical results.

Introduction

Hidden Markov models (HMMs) serve as a standard approach to modeling sequential data with underlying state changes and temporal dependence [9, 52, 21, 12, 65]. In this framework, the hidden states evolve according to a finite-state Markov chain, and each observation is generated from the emission distribution associated with its current state. Owing to these modeling capabilities, HMMs and closely related Markov-switching models have been employed extensively in speech recognition [52, 24], genomics [22, 30, 45], single-molecule biophysics [47, 11, 59], ecological modeling [36, 46], neuroscience [6, 61], and a broad range of time-series applications [23, 65]. In many of these applications, the central goal is to interpret the underlying process through fitted states and transitions. However, a close fit to the observations alone is insufficient for reliable interpretation unless the fitted emission distributions, state probabilities, and transition flows also accurately reflect their true counterparts. For Gaussian emissions with a common, known covariance, this difficulty becomes especially pronounced when emission centers approach one another. Specifically, models with substantially different state probabilities or transition matrices can produce nearly identical observation densities.

This gap between observation fit and latent recovery calls for deriving convergence rates that explicitly account for emission separation. To obtain these rates, we establish inverse inequalities that bound latent recovery error in terms of observation-density error and emission separation. Formally, let θ\theta denote a candidate HMM parameter, θ⋆\theta^\star the true parameter, and pθ,Tp_{\theta,T} the density of a length-TT observation sequence under θ\theta. Our inverse inequalities take the following form

Dr(θ,θ⋆)≤CrΔsep−αrh(pθ,T,pθ⋆,T)νr,(1)\mathcal{D}_r(\theta,\theta^\star) \le C_r \Delta_{\mathrm{sep}}^{-\alpha_r} h(p_{\theta,T},p_{\theta^\star,T})^{\nu_r}, \tag*{(1)}

where hh denotes the Hellinger distance defined in equation (14), Δsep\Delta_{\mathrm{sep}} is the truncated minimum separation between true emission centers, as defined in § 2.1, and Dr\mathcal{D}_r measures error in a latent target involving rr time coordinates. The constant CrC_r is independent of Δsep\Delta_{\mathrm{sep}}, while the exponents αr\alpha_r and νr\nu_r quantify the dependence of the recovery bound on emission separation and observation-density error, respectively. Once the Hellinger error of an estimator is controlled, these inequalities translate its density estimation rate into latent recovery rates. The recovery targets considered in this paper are model parameters and unconditional distributions of latent emission-center sequences, rather than the realized hidden path or its posterior distribution given the observations.

The separation exponent αr\alpha_r in (1) also depends on how many true emission centers approach one another, which is not captured by their minimum separation alone. We illustrate this dependence in Figure 1, where four emission centers form either two separated pairs or a single cluster, with the same minimum separation in both configurations. In the first configuration, the four centers form two pairs of nearby centers that remain well separated from each other, whereas in the second, all four centers lie close together in a single cluster. This difference changes the separation factor in the exact-specification emission-center bound from Δsep−2\Delta_{\mathrm{sep}}^{-2} to Δsep−6\Delta_{\mathrm{sep}}^{-6}, as established in Theorem 3.1. Driven by this example, we analyze how the number of nearby centers, in addition to their minimum separation, determines the sensitivity of latent recovery to observation error.

Two configurations of Gaussian emission distributions illustrating different separation exponentsVector (PDF)
Dμ≤C(πmin⁡∗)Δsep−(2smax⁡−2)h(pθ,T,pθ∗,T)D_\mu\leq C(\pi_{\min}^{*})\Delta_{\mathrm{sep}}^{-(2s_{\max}-2)}h(p_\theta,T,p_{\theta^*},T)

(a) Two separated pairs: K=4K=4, smax⁡=2s_{\max}=2, Dμ≲Δsep−2hD_\mu\lesssim\Delta_{\mathrm{sep}}^{-2}h \hfill (b) One cluster of four: K=4K=4, smax⁡=4s_{\max}=4, Dμ≲Δsep−6hD_\mu\lesssim\Delta_{\mathrm{sep}}^{-6}h

Figure 1. Same state count and minimum separation, different separation exponents. Both configurations have K=4K=4 Gaussian emission distributions and the same minimum separation Δsep\Delta_{\mathrm{sep}}. Colored dots mark emission centers, and dashed outlines group centers in the same cluster. Configuration (a) consists of two well-separated pairs with smax⁡=2s_{\max}=2, whereas configuration (b) consists of one cluster of four with smax⁡=4s_{\max}=4. Theorem 3.1 gives separation factors Δsep−2\Delta_{\mathrm{sep}}^{-2} and Δsep−6\Delta_{\mathrm{sep}}^{-6} in the global emission-center recovery bounds under exact specification for configurations (a) and (b), respectively.

Latent Recovery in Hidden Markov Models

Latent recovery in HMMs has long been a subject of statistical research, with classical work on likelihood-based estimation and model identification [21, 12]. Over time, this literature has expanded to include finite-sample recovery guarantees, with more recent work examining how emission separation affects estimation rates.

Estimation and identifiability. Likelihood-based inference provides a classical approach to estimating HMM parameters from observed sequences. Consistency and asymptotic normality describe the behavior of these estimators as the sample size increases under suitable assumptions [9, 42, 10, 19, 20], while Lehéricy [41] studies nonasymptotic guarantees for maximum likelihood under model misspecification. In practical applications, likelihood-based estimation also requires algorithms for computing the estimators. The expectation-maximization (EM) algorithm provides an iterative approach to this computation with latent variables [17], and Balakrishnan et al. [7] develop guarantees that connect its optimization and statistical errors under suitable initialization and local stability conditions. Apart from computation, the specification of the fitted model requires choosing the number of hidden states. When this number is unknown, state-count estimation is addressed through penalized likelihood and information-theoretic procedures ([25]) and nonparametric order estimators ([40]). From a Bayesian perspective, posterior analyses also account for an unknown state count or flexible emission distributions ([26, 60]).

Underlying these estimation approaches is the question of whether the latent parameters can be uniquely determined from the observation distribution. Since hidden-state labels are arbitrary, uniqueness is understood up to a common state permutation. General identifiability results for latent-variable models, including HMMs, are developed by Allman et al. [4] through algebraic arguments based on conditional independence. More specifically, Gassiat et al. [27] establish identifiability for finite-state nonparametric HMMs with a full-rank transition matrix and linearly independent emission distributions. The linear-independence requirement can be relaxed to distinct emissions when the transition matrix is full-rank and ergodic, as shown by Alexandrovich et al. [3]. These results clarify when the observation distribution distinguishes different latent models, but do not by themselves quantify how a small error in that distribution affects latent recovery. This quantitative relation is needed for recovery guarantees from finitely many observations, where the observation distribution must itself be estimated.

Finite-sample recovery. Spectral and method-of-moments approaches provide finite-sample guarantees by using observable moments to construct matrix or tensor decompositions of the latent model ([34, 54, 35, 5]). Building on these approaches, subsequent work improves the statistical efficiency of moment-based estimation ([57]) and extends spectral inference to hidden semi-Markov models ([49]). The recovery guarantees for such procedures depend on how accurately the latent quantities can be reconstructed from estimated moments. In the spectral analysis of Hsu et al. [34], this reconstruction is controlled through lower bounds for singular values of the model matrices. These conditions ensure that errors in observable quantities do not become arbitrarily amplified during recovery. The propagation of estimation error can also be studied directly through inverse inequalities that bound latent recovery error in terms of observation-density error. In the nonparametric HMM setting, Castro et al. [13] use this approach to derive emission-estimation rates. Specifically, they first establish an oracle inequality for estimating the joint density of three consecutive observations, then show under an additional nondegeneracy condition that the squared emission-recovery error is bounded by the squared joint-density error. This relation converts their joint-density guarantee into minimax-adaptive rates for emission estimation, up to logarithmic factors. However, these rates depend on the lowest regularity among the emission densities, so they do not give faster rates for smoother densities within the same model. Lehéricy [39] addresses this limitation through state-specific rates that adapt to the regularity of each emission density. In addition to emission-density estimation, parameter-recovery guarantees also support inference on hidden states conditional on the observations. For conditional inference, De Castro et al. [16] control filtering and marginal smoothing errors in terms of parameter-estimation errors and provide bounds for nonparametric spectral estimators. Nevertheless, the finite-sample results discussed above do not explicitly quantify how latent recovery rates depend on emission separation.

Recovery as emissions approach one another. Abraham et al. [1] explicitly quantify how parameter-estimation rates depend on emission separation in two-state HMMs with multinomial emissions. More precisely, they establish matching nonasymptotic minimax upper and lower bounds over classes in which the separation between the two emission probability vectors may decrease with the sample size. Consequently, their bounds quantify the additional data needed for recovery as the emissions become harder to distinguish. With nonparametric emissions, recovery additionally requires estimating the emission densities themselves, so their smoothness also affects estimation accuracy. Moreover, temporal dependence can provide information for distinguishing the hidden states, so recovery becomes more difficult as the observations approach an i.i.d. sequence. In this setting, Abraham et al. [2] obtain minimax bounds for two-state HMMs that account jointly for emission smoothness and how close the observations are to independence. Despite these differences in emission assumptions, both studies consider two-state HMMs, where separation involves only a single pair of emission distributions. With more states, however, the same minimum separation can arise from one cluster of approaching Gaussian emission centers or several smaller, separated clusters, as illustrated in Figure 1. Thus, the two-state results leave unresolved how the number of approaching centers within each cluster affects latent recovery rates.

Connections to Finite-Mixture Theory

For Gaussian HMMs with an arbitrary number of hidden states, finite-mixture theory provides a natural starting point for studying latent recovery. Indeed, over a fixed trajectory length, the joint observation density is a finite Gaussian mixture whose mixing measure describes the distribution of latent emission-center sequences. Therefore, recovering this measure amounts to recovering the latent location distributions of interest.

Mixing-distribution estimation. The mixture representation above connects latent recovery in Gaussian HMMs to the identification and estimation of mixing distributions. Both problems have been studied in finite and infinite mixture models [55, 64, 37, 8, 38], with broader discussions of mixture inference in [56, 43, 48]. At the population level, identifiability establishes when the observation density uniquely determines the mixing distribution. With finitely many observations, estimation additionally requires quantifying how accurately that distribution can be recovered from data. Likelihood-based methods address this estimation problem, and their computation is reviewed by Redner and Walker [53]. However, their statistical behavior can depart from regular parametric theory near singular or nonidentifiable configurations, as studied by Chen [14]; Dacunha-Castelle and Gassiat [15]; Liu and Shao [44]. Near such configurations, differences in component locations and probabilities can produce much smaller differences in the observation density. As a result, accurate density estimation need not imply equally accurate recovery of the mixing distribution. On the density-estimation side, Ghosal and van der Vaart [28] derive Hellinger convergence rates for maximum-likelihood estimators of Gaussian mixtures through bracketing-entropy bounds and empirical process tools (van de Geer, [58]). These guarantees control the observation-density error, but do not by themselves bound errors in component locations and probabilities. Accordingly, latent recovery requires an inverse bound that controls both types of error in terms of observation-density error.

Wasserstein recovery and separation geometry. Wasserstein distance measures errors in both component locations and probability masses, making it suitable for assessing mixing-distribution recovery (Villani, [62]). As a link between density estimation and mixing-distribution recovery, Nguyen [51] establishes quantitative relations between Wasserstein distances on mixing measures and divergences between the corresponding mixture densities for independent observations. Within this framework, the recovery rates depend on the identifiability properties of the mixture family and on whether the fitted model contains excess components. In particular, Ho and Nguyen [32] show how strong and weak identifiability lead to different convergence behavior in exactly specified and overfitted finite mixtures. At the local minimax level, Heinrich and Kahn [31] further establish how the number of excess components determines the optimal recovery rate near a fixed mixing distribution.

Through moment comparisons, Wasserstein recovery bounds can also be obtained for Gaussian location mixtures. Specifically, Wu and Yang [63] derive polynomial interpolation inequalities that convert moment discrepancies into Wasserstein bounds and use them to obtain optimal estimation guarantees. In higher dimensions, Doss et al. [18] exploit moment tensors to establish minimax-optimal Wasserstein rates without separation assumptions. These minimax guarantees describe the worst-case difficulty over classes allowing arbitrarily close components, rather than the dependence of recovery accuracy on a particular configuration of component locations. From a complementary perspective, Nguyen et al. [50] develop polynomial constructions that make the dependence on component separation and minimum mixture weight explicit in recovery bounds.

However, directly applying generic mixture bounds to latent recovery in Gaussian HMMs poses two distinct difficulties. First, component-count-dependent bounds can make recovery guarantees depend on up to KTK^T hidden paths, even though these paths reuse the same KK emission centers across time. Furthermore, even within this shared set, the total number of centers fails to reflect whether they all approach one another or instead concentrate into smaller, well-separated clusters. Second, unlike the freely varying weights inherent to generic mixtures, HMM path weights are constrained by the Markov factorization into initial-state and transition probabilities. In turn, generic bounds for unrestricted weights leave these structural dependencies unexploited.

Our Contributions

This paper addresses these difficulties through inverse inequalities that exploit the shared emission centers and Markov structure of Gaussian HMMs. These inequalities connect observation-density error to latent recovery error under both exact specification, where the fitted and true models have the same number of states, and over-specification, where the fitted model has more states than the true model. Combined with density-estimation guarantees, they yield statistical recovery rates that quantify how estimation accuracy depends on sample size, emission separation, and cluster size. Our main contributions are summarized as follows.

  1. HMM-aware inverse bounds for latent recovery. We establish deterministic inverse inequalities that bound latent recovery errors in terms of the Hellinger distance between fitted and true observation densities. Our analysis exploits the shared emission centers across time and the Markov factorization of path probabilities, avoiding a treatment of the observation density as an arbitrary KTK^T-component mixture. This structure allows us to express the dependence on emission separation through explicit powers, with constants independent of that separation. When the true centers form clusters whose diameters are bounded by a fixed multiple of the minimum separation and whose intercluster distances remain bounded away from zero, these powers depend on the largest cluster size smax⁡s_{\max}. Without cluster assumptions, the corresponding bounds hold with smax⁡=Ks_{\max}=K.

  1. Sharp separation exponents under exact specification and over-specification. We derive separation exponents under both exact specification and over-specification and prove their sharpness through explicit examples, with the powers of Hellinger error held fixed. Under exact specification, our global bounds have separation exponents 2smax⁡−12s_{\max}-1 for state-sequence probability recovery and 2smax⁡−22s_{\max}-2 for 1-Wasserstein recovery of joint emission-center distributions, neither of which increases with the number of selected times. The constants in these global bounds may depend on the minimum true initial-state probability. When the observation-density error is sufficiently small, we prove that fitted and true emission centers admit a one-to-one matching and obtain inverse bounds with the same separation exponents but constants independent of that probability. Under over-specification, however, several fitted states may represent the same true state, so their individual transition rows need not be identifiable. Therefore, we control probabilities aggregated over these states, with separation exponents that now depend on the number of selected times. We further extend these aggregate probability bounds to grouping based on sufficiently accurate reference centers without changing their separation exponents.

  1. Statistical recovery rates with explicit separation dependence. We combine a nearly parametric Hellinger bound for the global MLE with the inverse inequalities above to obtain latent recovery rates that depend explicitly on emission separation and cluster size. For instance, emission-center displacement, weighted by fitted state probabilities, has rate OP(Δsep−(2smax⁡−2)(d/n)1/2)\mathcal{O}_{\mathbb{P}}\left(\Delta_{\mathrm{sep}}^{-(2s_{\max}-2)}(d/n)^{1/2}\right) under exact specification and OP(Δsep−(smax⁡−1)(d/n)1/4)\mathcal{O}_{\mathbb{P}}\left(\Delta_{\mathrm{sep}}^{-(s_{\max}-1)}(d/n)^{1/4}\right) under over-specification, up to logarithmic factors for fixed trajectory length. In both cases, the separation factors increase as the true emission centers approach one another, so a larger sample size is needed to obtain the same upper bound. At the same minimum emission separation, these recovery rates have smaller inverse powers of separation when fewer emission centers approach one another within each cluster. This results in different statistical recovery guarantees for HMMs with the same state count and minimum emission separation, according to how many emission centers approach one another within the largest cluster.

In this paper, we consider independent trajectories observed over a fixed time horizon. Such settings arise in longitudinal studies and repeated experiments, where each subject or experimental run is observed at a fixed number of time points. In both cases, the amount of data increases through the number of trajectories rather than their length. As more independent trajectories become available, our recovery rates describe how accurately individual-state and aggregate latent quantities can be estimated, providing a statistical basis for interpreting fitted HMMs in applications such as disease progression, human activity recognition, and economic modeling.

Paper Organization. The remainder of the paper is organized as follows. § 2 introduces the model, latent recovery losses, emission separation and clustering conditions, and establishes the density-estimation bound used to obtain statistical recovery rates. With these definitions, we develop the main recovery bounds under exact specification and over-specification in § 3 and § 4, respectively. We empirically assess our recovery bounds and their corresponding statistical rates through synthetic experiments under exact specification and over-specification in § 5. § 6 discusses what these guarantees imply for the interpretation and use of fitted HMMs, and § 7 concludes the paper. Proofs, supporting technical lemmas, and additional results are provided in the appendices.

Notation and Preliminaries

In this section, we first introduce the Gaussian HMM and define loss functions for comparing fitted and true emission centers, state probabilities, transition flows, and emission-center distributions. Subsequently, we specify conditions on the separation of true emission centers within and between clusters that determine how observation-density error translates into latent recovery bounds. Finally, we establish a finite-sample Hellinger bound for the global maximum likelihood estimator, which we use in later sections to derive latent recovery rates from the inverse inequalities.

Gaussian Hidden Markov Models

Model parameters. Let K≥2K \ge2 be the number of true hidden states and let k≥Kk \ge K be the number of states in the fitted model. A Gaussian HMM is parameterized by θ=(π,A,μ1,…,μk)\theta= (\pi, A, \mu_1, \ldots, \mu_k), where π\pi is the row vector of initial state probabilities, AA is the transition matrix, and μ1,…,μk∈Rd\mu_1, \ldots, \mu_k \in\mathbb{R}^d are the emission centers. We write Pθ\mathbb{P}_{\theta} for probabilities under the model with parameter θ\theta. The latent states form a Markov chain with initial probability π\pi and transition matrix AA

Z1∼π,Pθ(Zt+1=j∣Zt=i)=Aij.Z_1 \sim\pi,\qquad\mathbb{P}_{\theta}(Z_{t+1}=j \mid Z_t=i) = A_{ij}.

Conditioned on the latent states, the observations are independent and satisfy Xt∣Z1:T∼Nd(μZt,Σ)X_t \mid Z_{1:T} \sim\mathcal{N}_d(\mu_{Z_t}, \Sigma), where the common covariance matrix Σ\Sigma is known and positive definite.

Observed data. We observe nn independent trajectories X1:T(1),…,X1:T(n)X_{1:T}^{(1)}, \ldots, X_{1:T}^{(n)} generated by the same Gaussian HMM with true parameter θ∗=(π∗,A∗,μ1∗,…,μK∗)\theta^* = (\pi^*, A^*, \mu_1^*, \ldots, \mu_K^*). Each trajectory consists of TT observations in Rd\mathbb{R}^d with its hidden-state sequence unobserved. Within each trajectory, however, the observations are dependent through the hidden Markov chain. Throughout our statistical analysis, T≥2T \ge2 remains fixed as nn tends to infinity, so additional data come from more independent trajectories rather than longer sequences.

Normalization of the emission covariance. For simplicity, we reduce the emission covariance to the identity matrix by letting S=Σ−1/2S = \Sigma^{-1/2} and transforming each observation XtX_t into SXtSX_t. Accordingly, the transformed emission density is φd(x−SμZt)\varphi_d(x - S\mu_{Z_t}), where φd\varphi_d denotes the standard Gaussian density on Rd\mathbb{R}^d. The initial distribution π\pi and transition matrix AA remain unchanged. This transformation also preserves the Hellinger distance in equation (1). For any u∈Rdu \in\mathbb{R}^d,

λmin⁡(Σ)∥Su∥2≤∥u∥2≤λmax⁡(Σ)∥Su∥2,\sqrt{\lambda_{\min}(\Sigma)}\lVert Su\rVert_2 \le\lVert u\rVert_2 \le\sqrt{\lambda_{\max}(\Sigma)}\lVert Su\rVert_2,

where λmin⁡(Σ)\lambda_{\min}(\Sigma) and λmax⁡(Σ)\lambda_{\max}(\Sigma) denote the smallest and largest eigenvalues of Σ\Sigma, respectively. For fixed Σ\Sigma, these inequalities imply that distances between emission centers before and after the transformation differ by at most constant factors depending on Σ\Sigma. Consequently, the recovery bounds established in subsequent sections for the transformed observations and emission centers can be expressed in the original coordinates without changing their orders. We provide the detailed justification of the reduction to identity emission covariance and its effect on the recovery bounds in Appendix § 9.§ 1.

Throughout the sequel, we retain the notation XtX_t and μj\mu_j for the transformed observations and centers. All radius constraints, separation and clustering conditions, and nearest-center assignments below are understood in these transformed coordinates.

Observation density. We write pθ,Tp_{\theta,T} for the density of a length-TT observation sequence under parameter θ\theta. For every positive integer mm, write [m]={1,…,m}[m] = \{1,\ldots,m\}. For any latent state sequence z∈[k]Tz \in[k]^T, its probability is

Pθ(Z1:T=z)=πz1∏t=1T−1Aztzt+1,\mathbb{P}_{\theta}(Z_{1:T}=z) = \pi_{z_1}\prod_{t=1}^{T-1} A_{z_tz_{t+1}},

by the Markov property. Conditional on Z1:T=zZ_{1:T}=z, the observations are independent, with joint density

pθ(x1:T∣Z1:T=z)=∏t=1Tφd(xt−μzt).p_{\theta}(x_{1:T}\mid Z_{1:T}=z) = \prod_{t=1}^{T}\varphi_d(x_t-\mu_{z_t}).

The unconditional density of the length-TT observation sequence X1:TX_{1:T} is obtained by summing the conditional densities over all latent paths z∈[k]Tz \in[k]^T, weighted by their respective probabilities:

References

  1. [1]Kweku Abraham, Elisabeth Gassiat, and Zacharie Naulet. Fundamental limits for learning hidden Markov model parameters. IEEE Transactions on Information Theory, 69(3):1777–1794, 2023.DOI
  2. [2]Kweku Abraham, Elisabeth Gassiat, and Zacharie Naulet. Frontiers to the learning of nonparametric hidden Markov models. Journal of Machine Learning Research, 26(155):1–75, 2025. URL http://jmlr.org/papers/v26/24-2230.html.
  3. [3]Grigory Alexandrovich, Hajo Holzmann, and Anna Leister. Nonparametric identification and maximum likelihood estimation for hidden Markov models. Biometrika, 103(2):423–434, 2016.pubmed.ncbi.nlm.nih.gov/27279667
  4. [4]Elizabeth S. Allman, Catherine Matias, and John A. Rhodes. Identifiability of parameters in latent structure models with many observed variables. The Annals of Statistics, 37(6A):3099 – 3132, 2009.DOI
  5. [5]Animashree Anandkumar, Rong Ge, Daniel Hsu, Sham M. Kakade, and Matus Telgarsky. Tensor decompositions for learning latent variable models. Journal of Machine Learning Research, 15(80):2773–2832, 2014.
  6. [6]Adam P Baker, Matthew J Brookes, Iead A Rezek, Stephen M Smith, Timothy Behrens, Penny J Probert Smith, and Mark Woolrich. Fast transient networks in spontaneous human brain activity. eLife, 3:e01867, 2014.DOI
  7. [7]Sivaramam Balakrishnan, Martin J. Wainwright, and Bin Yu. Statistical guarantees for the EM algorithm: From population to sample-based analysis. The Annals of Statistics, 45(1):77 – 120, 2017.arxiv.org/abs/1408.2156
  8. [8]Nicola Bariletto, Dung Le, Alessandro Rinaldo, and Nhat Ho. Convergence rates for latent mixing measures in infinite homoscedastic location-scale mixture models. arXiv preprint arXiv:2605.06845, 2026.arxiv.org/abs/2605.06845
  9. [9]Leonard E Baum and Ted Petrie. Statistical inference for probabilistic functions of finite state Markov chains. The Annals of Mathematical Statistics, 37(6):1554–1563, 1966.DOI
  10. [10]Peter J Bickel, Ya’acov Ritov, and Tobias Rydén. Asymptotic normality of the maximum-likelihood estimator for general hidden Markov models. The Annals of Statistics, 26(4):1614–1635, 1998.DOI
  11. [11]Jonathan E Bronson, Jingyi Fei, Jake M Hofman, Ruben L Gonzalez, and Chris H Wiggins. Learning rates and states from biophysical time series: A Bayesian approach to model selection and single-molecule FRET data. Biophysical Journal, 97(12):3196–3205, 2009.DOI
  12. [12]Olivier Cappé, Eric Moulines, and Tobias Rydén. Inference in hidden Markov models. Springer, 2005.DOI
  13. [13]Yohann De Castro, Élisabeth Gassiat, and Claire Lacour. Minimax adaptive estimation of nonparametric hidden Markov models. Journal of Machine Learning Research, 17(111):1–43, 2016. URL http://jmlr.org/papers/v17/15-381.html.arxiv.org/abs/1501.04787
  14. [14]Jiahua Chen. Optimal rate of convergence for finite mixture models. The Annals of Statistics, 23(1):221 – 233, 1995.DOI
  15. [15]D. Dacunha-Castelle and E. Gassiat. Testing the order of a model using locally conic parametrization: Population mixtures and stationary ARMA processes. The Annals of Statistics, 27(4):1178 – 1209, 1999.
  16. [16]Yohann De Castro, Élisabeth Gassiat, and Sylvain Le Corff. Consistent estimation of the filtering and marginal smoothing distributions in nonparametric hidden Markov models. IEEE Transactions on Information Theory, 63(8):4758–4777, 2017.arxiv.org/abs/1507.06510
  17. [17]Arthur P Dempster, Nan M Laird, and Donald B Rubin. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society: Series B (Methodological), 39(1):1–22, 1977.DOI
  18. [18]Natalie Doss, Yihong Wu, Pengkun Yang, and Harrison H. Zhou. Optimal estimation of high-dimensional Gaussian location mixtures. The Annals of Statistics, 51(1):62 – 95, 2023.
  19. [19]Randal Douc, Éric Moulines, and Tobias Rydén. Asymptotic properties of the maximum likelihood estimator in autoregressive models with Markov regime. The Annals of Statistics, 32(5):2254–2304, 2004.
  20. [20]Randal Douc, Eric Moulines, Jimmy Olsson, and Ramon van Handel. Consistency of the maximum likelihood estimator for general hidden Markov models. The Annals of Statistics, 39(1):474–513, 2011.
  21. [21]Yariv Ephraim and Neri Merhav. Hidden Markov processes. IEEE Transactions on Information Theory, 48(6):1518–1569, 2002.DOI
  22. [22]Jane Fridlyand, Antoine M Snijders, Dan Pinkel, Donna G Albertson, and Ajay N Jain. Hidden Markov models approach to the analysis of array CGH data. Journal of Multivariate Analysis, 90 (1):132–153, 2004.DOI
  23. [23]Sylvia Frühwirth-Schnatter. Finite mixture and Markov switching models. Springer, 2006.
  24. [24]Mark J. F. Gales and Steve J. Young. The application of hidden Markov models in speech recognition. Found. Trends Signal Process., 1(3):195–304, 2007.
  25. [25]Élisabeth Gassiat and Stéphane Boucheron. Optimal error exponents in hidden Markov models order estimation. IEEE Transactions on Information Theory, 49(4):964–980, 2003.DOI
  26. [26]Élisabeth Gassiat and Judith Rousseau. About the posterior distribution in hidden Markov models with unknown number of states. Bernoulli, 20(4):2039 – 2075, 2014.DOI
  27. [27]Élisabeth Gassiat, Alice Cleynen, and Stephane Robin. Inference in finite state space non parametric hidden Markov models and applications. Statistics and Computing, 26:61–71, 2016.DOI
  28. [28]Subhashis Ghosal and Aad W. van der Vaart. Entropies and rates of convergence for maximum likelihood and Bayes estimation for mixtures of normal densities. The Annals of Statistics, 29(5):1233 – 1263, 2001.
  29. [29]Gene H Golub and John H Welsch. Calculation of Gauss quadrature rules. Mathematics of Computation, 23(106):221–230, 1969.DOI
  30. [30]Subharup Guha, Yi Li, and Donna Neuberg. Bayesian hidden Markov modeling of array CGH data. Journal of the American Statistical Association, 103(482):485–497, 2008.
  31. [31]Philippe Heinrich and Jonas Kahn. Strong identifiability and optimal minimax rates for finite mixture estimation. The Annals of Statistics, 46(6A):2844 – 2870, 2018. doi: 10.1214/17-AOS1641. URL https://doi.org/10.1214/17-AOS1641.
  32. [32]Nhat Ho and XuanLong Nguyen. Convergence rates of parameter estimation for some weakly identifiable finite mixtures. The Annals of Statistics, 44(6):2726 – 2755, 2016a.DOI
  33. [34]Daniel Hsu, Sham M. Kakade, and Tong Zhang. A spectral algorithm for learning hidden Markov models. Journal of Computer and System Sciences, 78(5):1460–1480, 2012.arxiv.org/abs/0811.4413
  34. [35]Tzu-Kuo Huang and Jeff Schneider. Spectral learning of hidden Markov models from dynamic and static data. In Proceedings of the 30th International Conference on Machine Learning, volume 28 of Proceedings of Machine Learning Research, pages 630–638. PMLR, 2013.
  35. [36]Roland Langrock, Ruth King, Jason Mattiopoulos, Len Thomas, Daniel Fortin, and Juan M Morales. Flexible and practical modeling of animal telemetry data: hidden Markov models and extensions. Ecology, 93(11):2336–2342, 2012.
  36. [37]Dung Le, Nicola Bariletto, Alessandro Rinaldo, and Nhat Ho. Partial differential equation barriers to identifiability in infinite mixture models. arXiv preprint arXiv:2608.08597, 2026a.arxiv.org/abs/2608.08597
  37. [38]Dung Le, Huy Nguyen, Trang Pham, Alessandro Rinaldo, and Nhat Ho. Characterizing heterogeneous rates in finite mixture estimation via partial optimal transport. arXiv preprint arXiv:2609.16622, 2026b.arxiv.org/abs/2609.16622
  38. [39]Luc Lehéřicy. State-by-state minimax adaptive estimation for nonparametric hidden Markov models. Journal of Machine Learning Research, 19(39):1–46, 2018.arxiv.org/abs/1706.08277
  39. [40]Luc Lehéřicy. Consistent order estimation for nonparametric hidden Markov models. Bernoulli, 25 (1):464–498, 2019.arxiv.org/abs/1606.00622
  40. [41]Luc Lehéřicy. Nonasymptotic control of the MLE for misspecified nonparametric hidden Markov models. Electronic Journal of Statistics, 15(2):4916–4965, 2021.arxiv.org/abs/1807.03997
  41. [42]Brian G. Leroux. Maximum-likelihood estimation for hidden Markov models. Stochastic Processes and their Applications, 40(1):127–143, 1992.DOI
  42. [43]Bruce G. Lindsay. Mixture models: Theory, geometry and applications. NSF-CBMS Regional Conference Series in Probability and Statistics, 5:1–163, 1995.DOI
  43. [44]Xin Liu and Yongzhao Shao. Asymptotics for likelihood ratio tests under loss of identifiability. The Annals of Statistics, 31(3):807 – 832, 2003.DOI
  44. [45]Md Pavel Mahmud and Alexander Schliep. Fast MCMC sampling for hidden Markov models to determine copy number variations. BMC Bioinformatics, 12(1):428, 2011.DOI
  45. [46]Brett T McClintock, Roland Langrock, Olivier Gimenez, Emmanuelle Cam, David L Borchers, Richard Glennie, and Toby A Patterson. Uncovering ecological state dynamics with hidden Markov models. Ecology Letters, 23(12):1878–1903, 2020.DOI
  46. [47]Sean A McKinney, Chirlmin Joo, and Taekjip Ha. Analysis of single-molecule FRET trajectories using hidden Markov modeling. Biophysical Journal, 91(5):1941–1951, 2006.pubmed.ncbi.nlm.nih.gov/16766620
  47. [48]Geoffrey McLachlan and David Peel. Finite mixture models. Wiley Series in Probability and Statistics. John Wiley & Sons, 2000.DOI
  48. [49]Igor Melnyk and Arindam Banerjee. A spectral algorithm for inference in hidden semi-Markov models. Journal of Machine Learning Research, 18(35):1–39, 2017.
  49. [50]Huy Nguyen, Dung Le, Alessandro Rinaldo, and Nhat Ho. On the geometry of separation in finite Gaussian mixtures. arXiv preprint arXiv:2606.16179, 2026.arxiv.org/abs/2606.16179
  50. [51]XuanLong Nguyen. Convergence of latent mixing measures in finite and infinite mixture models. The Annals of Statistics, 41(1):370 – 400, 2013.
  51. [52]Lawrence R Rabiner. A tutorial on hidden Markov models and selected applications in speech recognition. Proceedings of the IEEE, 77(2):257–286, 1989.
  52. [53]Richard A. Redner and Homer F. Walker. Mixture densities, maximum likelihood and the EM algorithm. SIAM Review, 26(2):195–239, 1984.
  53. [54]Sajid Siddiqi, Byron Boots, and Geoffrey Gordon. Reduced-rank hidden Markov models. In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, volume 9 of Proceedings of Machine Learning Research, pages 741–748. PMLR, 2010.arxiv.org/abs/0910.0902
  54. [55]Henry Teicher. Identifiability of finite mixtures. The Annals of Mathematical Statistics, 34(4):1265 – 1269, 1963.DOI
  55. [56]D.M. Titterington, A.F.M. Smith, and U.E. Makov. Statistical analysis of finite mixture distributions. Wiley, New York, 1985.
  56. [57]Dustin Tran, Minjae Kim, and Finale Doshi-Velez. Spectral M-estimation with applications to hidden Markov models. In Proceedings of the 19th International Conference on Artificial Intelligence and Statistics, volume 51 of Proceedings of Machine Learning Research, pages 1421–1430. PMLR, 09–11 May 2016.arxiv.org/abs/1603.08815
  57. [58]Sara A. van de Geer. Empirical processes in M-estimation. Cambridge Series in Statistical and Probabilistic Mathematics. Cambridge University Press, 2000.
  58. [59]Jan-Willem van de Meent, Jonathan E Bronson, Chris H Wiggins, and Ruben L Gonzalez. Empirical Bayes methods enable advanced population-level analyses of single-molecule FRET experiments. Biophysical Journal, 106(6):1327–1337, 2014.DOI
  59. [60]Elodie Vernet. Posterior consistency for nonparametric hidden Markov models with finite state space. Electronic Journal of Statistics, 9(1):717 – 752, 2015.arxiv.org/abs/1311.3092
  60. [61]Diego Vidaurre, Stephen M Smith, and Mark W Woolrich. Brain network dynamics are hierarchically organized in time. Proceedings of the National Academy of Sciences, 114(48):12827–12832, 2017.DOI
  61. [62]Cédric Villani. Topics in optimal transportation. American Mathematical Society, 2003.DOI
  62. [63]Yihong Wu and Pengkun Yang. Optimal estimation of Gaussian mixtures via denoised method of moments. The Annals of Statistics, 48(4):1981 – 2007, 2020.arxiv.org/abs/1807.07237
  63. [64]Sidney J. Yakowitz and John D. Spragins. On the identifiability of finite mixtures. The Annals of Mathematical Statistics, 39(1):209 – 214, 1968.DOI
  64. [65]W. Zucchini, I.L. MacDonald, and R. Langrock. Hidden Markov models for time series: An introduction using R. Chapman and Hall/CRC Monographs on Statistics and Applied Probability Series. CRC Press, Taylor & Francis Group, Second edition, 2016.

Paper details

vlnc:W_vkeym7qw
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 6

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents