Efficient decoding with generative score-spaces using the expectation semiring

State-of-the-art speech recognisers are usually based on hidden Markov models (HMMs). They model a hidden symbol sequence with a Markov process, with the observations independent given that sequence. These assumptions yield efficient algorithms, but limit the power of the model. An alternative model that allows a wide range of features, including word- and phone-level features, is a log-linear model. To handle, for example, word-level variable-length features, the original feature vectors must be segmented into words. Thus, decoding must find the optimal combination of segmentation of the utterance into words and word sequence. Features must therefore be extracted for each possible segment of audio. For many types of features, this becomes slow. In this paper, long-span features are derived from the likelihoods of word HMMs. Derivatives of the log-likelihoods, which break the Markov assumption, are appended. Previously, decoding with this model took cubic time in the length of the sequence, and longer for higher-order derivatives. This paper shows how to decode in quadratic time. © 2013 IEEE.

Dynamics of filopodium-like protrusion and endothelial cellular motility on onedimensional extracellular matrix fibrils

Endothelial filopodia play key roles in guiding the tubular sprouting during angiogenesis. However, their dynamic morphological characteristics, with the associated implications in cell motility, have been subjected to limited investigations. In this work, the interaction between endothelial cells and extracellular matrix fibrils was recapitulated in vitro, where a specific focus was paid to derive the key morphological parameters to define the dynamics of filopodium-like protrusion during cell motility. Based on one-dimensional gelatin fibrils patterned by near-field electrospinning (NFES), we study the response of endothelial cells (EA.hy926) under normal culture or ROCK inhibition. It is shown that the behaviour of temporal protrusion length versus cell motility can be divided into distinct modes. Persistent migration was found to be one of the modes which permitted cell displacement for over 300 μm at a speed of approximately 1 μm min-1. ROCK inhibition resulted in abnormally long protrusions and diminished the persistent migration, but dramatically increased the speeds of protrusion extension and retraction. Finally, we also report the breakage of protrusion during cell motility, and examine its phenotypic behaviours. © 2014 The Author(s) Published by the Royal Society. All rights reserved.

Fixed-rank matrix factorizations and Riemannian low-rank optimization

Motivated by the problem of learning a linear regression model whose parameter is a large fixed-rank non-symmetric matrix, we consider the optimization of a smooth cost function defined on the set of fixed-rank matrices. We adopt the geometric framework of optimization on Riemannian quotient manifolds. We study the underlying geometries of several well-known fixed-rank matrix factorizations and then exploit the Riemannian quotient geometry of the search space in the design of a class of gradient descent and trust-region algorithms. The proposed algorithms generalize our previous results on fixed-rank symmetric positive semidefinite matrices, apply to a broad range of applications, scale to high-dimensional problems, and confer a geometric basis to recent contributions on the learning of fixed-rank non-symmetric matrices. We make connections with existing algorithms in the context of low-rank matrix completion and discuss the usefulness of the proposed framework. Numerical experiments suggest that the proposed algorithms compete with state-of-the-art algorithms and that manifold optimization offers an effective and versatile framework for the design of machine learning algorithms that learn a fixed-rank matrix. © 2013 Springer-Verlag Berlin Heidelberg.