Zero-Truncated Poisson Regression for Sparse Multiway Count Data Corrupted by False Zeros
arXiv:2201.10014 · doi:10.1093/imaiai/iaad016
Abstract
We propose a novel statistical inference methodology for multiway count data that is corrupted by false zeros that are indistinguishable from true zero counts. Our approach consists of zero-truncating the Poisson distribution to neglect all zero values. This simple truncated approach dispenses with the need to distinguish between true and false zero counts and reduces the amount of data to be processed. Inference is accomplished via tensor completion that imposes low-rank tensor structure on the Poisson parameter space. Our main result shows that an -way rank- parametric tensor generating Poisson observations can be accurately estimated by zero-truncated Poisson regression from approximately non-zero counts under the nonnegative canonical polyadic decomposition. Our result also quantifies the error made by zero-truncating the Poisson distribution when the parameter is uniformly bounded from below. Therefore, under a low-rank multiparameter model, we propose an implementable approach guaranteed to achieve accurate regression in under-determined scenarios with substantial corruption by false zeros. Several numerical experiments are presented to explore the theoretical results.
arXiv comment30 pages, 5 figures
References in corpus (10)
- BERTopic: Neural topic modeling with a class-based TF-IDF procedure
- Text Classification Algorithms: A Survey
- The Convex Geometry of Linear Inverse Problems
- On Tensors, Sparsity, and Nonnegative Factorizations
- Real-time 3D reconstruction from single-photon lidar data using plug-and-play point cloud denoisers
- Statistical detection of systematic election irregularities
- Generalized Canonical Polyadic Tensor Decomposition
- Poisson Matrix Recovery and Completion
- Stochastic Gradients for Large-Scale Tensor Decomposition
- Learning tensors from partial binary measurements