easily to address the problems when multiple observed tensors are factorised simultaneously. We illustrate our coupled factorisation approach on synthetic data as well as on a musical audio restoration problem. 1

of matrices and higher-order tensors. We propose to use an approach based on probabilistic interpretation of tensor factorisation models, i.e., Generalised Coupled Tensor Factorisation, which can simultaneously fit a large class of tensor models to higher-order tensors/matrices with com-mon latent factors

Generalised coupled tensor factorisation is a recently proposed algorithmic framework for simultaneously estimating tensor factorisation models where several observed tensors can share a set of latent factors. This paper proposes a model in this framework for coupled factorisation of piano

on such databases. In this paper, we extend FDB to support a larger class of practical queries with aggregates and ordering. This requires novel optimisation and evaluation techniques. We show how factorisation coupled with partial aggregation can effectively reduce the number of operations needed for query

The soft pomeron successfully correlates a wide variety of data. Its properties seem rather simple: it couples to single quarks and its coupling factorises. 1

We classify compactification lattices for supersymmetric Z2×Z2 orbifolds. These lattices include factorisable as well as non-factorisable six-tori. Different models lead to different numbers of fixed points/tori. A lower bound on the number of fixed tori per twisted sector is given by four, whereas

We study the anomalous dimensions and coefficient functions generated by the BFKL equation in 4 + 2ε dimensions, by investigating both running coupling effects, and the inclusion of the full next-to-leading kernel. After generalising the Fourier representation of the solutions to this case, we

between states, reduced performance in dynamic environments, and a lack of scalability. Factor-Q, the new algorithm, uses factorised state and action, coupled with multiple structured rewards, to address these issues. Initial experimental results demonstrate that Factor-Q is able to learn as efficiently

. A DML-HMM is built using Schwarz’s Bayesian Information Cri-terion based factorisation resulting in its topology being in-trinsically determined by the underlying causality and tem-poral order among different object events. Our experiments demonstrate that its performance on modelling group activ

-based speech enhancement. Index Terms — Non-negative matrix factorisation, coupled dic-tionary training, speech enhancement, modulation envelope 1.