In this paper, we propose a novel estimator for probabilistic models on discrete space, which is derived from an empirically localized homogeneous divergence.

The proposed estimator is based on un- normalized models and an empirically localized PS-divergence having the homogeneous property and can be constructed ...

To resolve this problem, we propose a novel estimator using ideas of a homogeneous divergence and an empirical localization in the following section. 3.

A novel estimator for probabilistic models on discrete space, which is derived from an empirically localized homogeneous divergence, and comparably performs ...

Jan 1, 2017 · In this paper, we propose a novel estimator for probabilistic models on discrete space, which is derived from an empirically localized ...

Sep 11, 2024 · Bibliographic details on Statistical Inference with Unnormalized Discrete Models and Localized Homogeneous Divergences.

A simple modification of the regularized covariance matrix estimator is proposed to make it PD while preserving the support, and is shown to preserve the ...

Jun 6, 2020 · Takenouchi, T. and T. Kanamori (2017). Statistical infer- ence with unnormalized discrete models and localized homogeneous divergences. J. Mach.

In this paper, we propose a novel parameter estimator for probabilistic models on discrete space. The proposed estimator is derived from minimization of ...

Statistical inference with unnormalized discrete models and localized homogeneous divergences ... Empirical localization of homogeneous divergences on discrete ...