Add to ORKG8 pages, 3 figures, 1 tableWe consider tensor factorizations using a generative model and a Bayesian approach. We compute rigorously the mutual information, the Minimal Mean Square Error (MMSE), and unveil information-theoretic phase transitions. In addition, we study the performance of Approximate Message Passing (AMP) and show that it achieves the MMSE for a large set of parameters, and that factorization is algorithmically "easy" in a much wider region than previously believed. It exists, however, a "hard" region where AMP fails to reach the MMSE and we conjecture that no polynomial algorithm will improve on AMP
International audienceIn this work we analyse quantitatively the interplay between the loss landscap...
Multiway data, described by tensors, are common in real-world applications. For example, online adve...
We describe an approach to speed-up inference with latent-variable PCFGs, which have been shown to b...
8 pages, 3 figures, 1 tableInternational audienceWe consider tensor factorizations using a generativ...
These notes present in a unified manner recent results (as well as new developments) on the informat...
In this thesis, we consider optimization problems that involve statistically estimating signals from...
We consider the Principal Component Analysis problem for large tensors of arbitrary order k under a ...
Abstract—We propose a generative model for robust tensor factorization in the presence of both missi...
Abstract. We present a Bayesian non-negative tensor factorization model for count-valued tensor data...
Graphical models (CRFs, Markov Random Fields, Bayesian networks ...) are probabilistic models fairly...
Tensor factorization is an important approach to multiway data analysis. How-ever, real-world tensor...
Factorizing low-rank matrices has many applications in machine learning and statistics. For probabil...
Because of the limitations of matrix factorization, such as losing spatial structure information, th...
Tensor factorization methods provide a useful way to extract latent factors from complex multirela-t...
International audienceIn this work we analyse quantitatively the interplay between the loss landscap...
Multiway data, described by tensors, are common in real-world applications. For example, online adve...
We describe an approach to speed-up inference with latent-variable PCFGs, which have been shown to b...
8 pages, 3 figures, 1 tableInternational audienceWe consider tensor factorizations using a generativ...
These notes present in a unified manner recent results (as well as new developments) on the informat...
In this thesis, we consider optimization problems that involve statistically estimating signals from...
We consider the Principal Component Analysis problem for large tensors of arbitrary order k under a ...
Abstract—We propose a generative model for robust tensor factorization in the presence of both missi...
Abstract. We present a Bayesian non-negative tensor factorization model for count-valued tensor data...
Graphical models (CRFs, Markov Random Fields, Bayesian networks ...) are probabilistic models fairly...
Tensor factorization is an important approach to multiway data analysis. How-ever, real-world tensor...
Factorizing low-rank matrices has many applications in machine learning and statistics. For probabil...
Because of the limitations of matrix factorization, such as losing spatial structure information, th...
Tensor factorization methods provide a useful way to extract latent factors from complex multirela-t...
International audienceIn this work we analyse quantitatively the interplay between the loss landscap...
Multiway data, described by tensors, are common in real-world applications. For example, online adve...
We describe an approach to speed-up inference with latent-variable PCFGs, which have been shown to b...