Corpus ID: 240420231

Factorization of Binary Matrices: Rank Relations, Uniqueness and Model Selection of Boolean Decomposition

  title={Factorization of Binary Matrices: Rank Relations, Uniqueness and Model Selection of Boolean Decomposition},
  author={Derek DeSantis and Erik West Skau and Duc P. Truong and Boian S. Alexandrov},
The application of binary matrices are numerous. Representing a matrix as a mixture of a small collection of latent vectors via low-rank decomposition is often seen as an advantageous method to interpret and analyze data. In this work, we examine the factorizations of binary matrices using standard arithmetic (real and nonnegative) and logical operations (Boolean and Z2). We examine the relationships between the different ranks, and discuss when factorization is unique. In particular, we… Expand

Figures and Tables from this paper


Boolean decomposition of binary matrices using a post-nonlinear mixture approach
This work proposes a simple yet efficient algorithm for solving the proposed BMF problem based on multiplicative update rules and provides for the first time in the binary data literature, a necessary and sufficient condition for the uniqueness of the Boolean matrix factorization, as well as several other uniqueness results. Expand
Modeling by shortest data
  • description. Automatica,
  • 1978
Boolean Matrix Factorization via Nonnegative Auxiliary Optimization
The proofs for the equivalencies of the two solution spaces under the existence of an exact solution of the BMF problem are provided, and the nonincreasing property of the algorithm is also proven. Expand
Finding the Number of Latent Topics With Semantic Non-Negative Matrix Factorization
This paper considers a semantic-assisted non-negative matrix factorization (NMF) topics model, which is based on Kullback-Leibler(KL) divergence and integrated with a method for determining the number of latent topics, and demonstrates that it not only determines the correct number of topics, but also extracts topics with a high coherence and accurately classifies the documents of the corpus. Expand
From-below Boolean matrix factorization algorithm based on MDL
This paper proposes a novel from-below Boolean matrix factorization algorithm based on formal concept analysis that outperforms—from different standpoints—existing state-of-the-art BMF algorithms. Expand
Model order selection for approximate Boolean matrix factorization problem
It is shown that the number of factors for DBP cannot be used in AFP, and a novel way is presented, reflecting the nature of AFP, how to establish the proper number of Factors. Expand
A neural network for determination of latent dimensionality in non-negative matrix factorization
A supervised machine learning approach in combination with a recent method for model determination, called NMFk, to determine the number of hidden features automatically, and compared the accuracy of the method to ARD, AIC and stability-based methods. Expand
Recent Developments in Boolean Matrix Factorization
A concise summary of the efforts of all of the communities studying Boolean Matrix Factorization is given and some open questions which in this opinion require further investigation are raised. Expand
A Practical Randomized CP Tensor Decomposition
The CANDECOMP/PARAFAC (CP) decomposition is a leading method for the analysis of multiway data and involves a series of alternating least squares algorithms. Expand
MDL4BMF: Minimum Description Length for Boolean Matrix Factorization
An existing algorithm for BMF is extended to use MDL to identify the best Boolean matrix factorization, analyze the complexity of the problem, and perform an extensive experimental evaluation to study its behavior. Expand