Streaming Bayesian Deep Tensor Factorization

Shikai Fang, Zheng Wang, Zhimeng Pan, Ji Liu, Shandian Zhe
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:3133-3142, 2021.

Abstract

Despite the success of existing tensor factorization methods, most of them conduct a multilinear decomposition, and rarely exploit powerful modeling frameworks, like deep neural networks, to capture a variety of complicated interactions in data. More important, for highly expressive, deep factorization, we lack an effective approach to handle streaming data, which are ubiquitous in real-world applications. To address these issues, we propose SBTD, a Streaming Bayesian Deep Tensor factorization method. We first use Bayesian neural networks (NNs) to build a deep tensor factorization model. We assign a spike-and-slab prior over each NN weight to encourage sparsity and to prevent overfitting. We then use multivariate Delta’s method and moment matching to approximate the posterior of the NN output and calculate the running model evidence, based on which we develop an efficient streaming posterior inference algorithm in the assumed-density-filtering and expectation propagation framework. Our algorithm provides responsive incremental updates for the posterior of the latent factors and NN weights upon receiving newly observed tensor entries, and meanwhile identify and inhibit redundant/useless weights. We show the advantages of our approach in four real-world applications.

Cite this Paper


BibTeX
@InProceedings{pmlr-v139-fang21d, title = {Streaming Bayesian Deep Tensor Factorization}, author = {Fang, Shikai and Wang, Zheng and Pan, Zhimeng and Liu, Ji and Zhe, Shandian}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {3133--3142}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/fang21d/fang21d.pdf}, url = {https://proceedings.mlr.press/v139/fang21d.html}, abstract = {Despite the success of existing tensor factorization methods, most of them conduct a multilinear decomposition, and rarely exploit powerful modeling frameworks, like deep neural networks, to capture a variety of complicated interactions in data. More important, for highly expressive, deep factorization, we lack an effective approach to handle streaming data, which are ubiquitous in real-world applications. To address these issues, we propose SBTD, a Streaming Bayesian Deep Tensor factorization method. We first use Bayesian neural networks (NNs) to build a deep tensor factorization model. We assign a spike-and-slab prior over each NN weight to encourage sparsity and to prevent overfitting. We then use multivariate Delta’s method and moment matching to approximate the posterior of the NN output and calculate the running model evidence, based on which we develop an efficient streaming posterior inference algorithm in the assumed-density-filtering and expectation propagation framework. Our algorithm provides responsive incremental updates for the posterior of the latent factors and NN weights upon receiving newly observed tensor entries, and meanwhile identify and inhibit redundant/useless weights. We show the advantages of our approach in four real-world applications.} }
Endnote
%0 Conference Paper %T Streaming Bayesian Deep Tensor Factorization %A Shikai Fang %A Zheng Wang %A Zhimeng Pan %A Ji Liu %A Shandian Zhe %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-fang21d %I PMLR %P 3133--3142 %U https://proceedings.mlr.press/v139/fang21d.html %V 139 %X Despite the success of existing tensor factorization methods, most of them conduct a multilinear decomposition, and rarely exploit powerful modeling frameworks, like deep neural networks, to capture a variety of complicated interactions in data. More important, for highly expressive, deep factorization, we lack an effective approach to handle streaming data, which are ubiquitous in real-world applications. To address these issues, we propose SBTD, a Streaming Bayesian Deep Tensor factorization method. We first use Bayesian neural networks (NNs) to build a deep tensor factorization model. We assign a spike-and-slab prior over each NN weight to encourage sparsity and to prevent overfitting. We then use multivariate Delta’s method and moment matching to approximate the posterior of the NN output and calculate the running model evidence, based on which we develop an efficient streaming posterior inference algorithm in the assumed-density-filtering and expectation propagation framework. Our algorithm provides responsive incremental updates for the posterior of the latent factors and NN weights upon receiving newly observed tensor entries, and meanwhile identify and inhibit redundant/useless weights. We show the advantages of our approach in four real-world applications.
APA
Fang, S., Wang, Z., Pan, Z., Liu, J. & Zhe, S.. (2021). Streaming Bayesian Deep Tensor Factorization. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:3133-3142 Available from https://proceedings.mlr.press/v139/fang21d.html.

Related Material