Hierarchical learning of grids of microtopics

Nebojsa Jojic, Alessandro Perina, Dongwoo Kim
Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, PMLR R14:148-157, 2016.

Abstract

The counting grid is a grid of microtopics, sparse word/feature distributions. The generative model associated with the grid does not use these microtopics individually, but in predefined groups which can only be (ad)mixed as such. Each allowed group corresponds to one of all possible overlapping rectangular windows into the grid. The capacity of the model is controlled by the ratio of the grid size and the window size.This paper builds upon the basic counting grid model and it shows that hierarchical reasoning helps avoid bad local minima, produces better classification accuracy and, most interestingly, allows for extraction of large numbers of coherent microtopics even from small datasets. We evaluate this in terms of consistency, diversity and clarity of the indexed content, as well as in a user study on word intrusion tasks. We demonstrate that these models work well as a technique for embedding raw images and discuss interesting parallels between hierarchical CG models and other deep architectures.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR14-jojic16a, title = {Hierarchical learning of grids of microtopics}, author = {Jojic, Nebojsa and Perina, Alessandro and Kim, Dongwoo}, booktitle = {Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence}, pages = {148--157}, year = {2016}, editor = {Ihler, Alexander and Janzing, Dominik}, volume = {R14}, series = {Proceedings of Machine Learning Research}, month = {25--29 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r14/main/assets/jojic16a/jojic16a.pdf}, url = {https://proceedings.mlr.press/r14/jojic16a.html}, abstract = {The counting grid is a grid of microtopics, sparse word/feature distributions. The generative model associated with the grid does not use these microtopics individually, but in predefined groups which can only be (ad)mixed as such. Each allowed group corresponds to one of all possible overlapping rectangular windows into the grid. The capacity of the model is controlled by the ratio of the grid size and the window size.This paper builds upon the basic counting grid model and it shows that hierarchical reasoning helps avoid bad local minima, produces better classification accuracy and, most interestingly, allows for extraction of large numbers of coherent microtopics even from small datasets. We evaluate this in terms of consistency, diversity and clarity of the indexed content, as well as in a user study on word intrusion tasks. We demonstrate that these models work well as a technique for embedding raw images and discuss interesting parallels between hierarchical CG models and other deep architectures.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Hierarchical learning of grids of microtopics %A Nebojsa Jojic %A Alessandro Perina %A Dongwoo Kim %B Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2016 %E Alexander Ihler %E Dominik Janzing %F pmlr-vR14-jojic16a %I PMLR %P 148--157 %U https://proceedings.mlr.press/r14/jojic16a.html %V R14 %X The counting grid is a grid of microtopics, sparse word/feature distributions. The generative model associated with the grid does not use these microtopics individually, but in predefined groups which can only be (ad)mixed as such. Each allowed group corresponds to one of all possible overlapping rectangular windows into the grid. The capacity of the model is controlled by the ratio of the grid size and the window size.This paper builds upon the basic counting grid model and it shows that hierarchical reasoning helps avoid bad local minima, produces better classification accuracy and, most interestingly, allows for extraction of large numbers of coherent microtopics even from small datasets. We evaluate this in terms of consistency, diversity and clarity of the indexed content, as well as in a user study on word intrusion tasks. We demonstrate that these models work well as a technique for embedding raw images and discuss interesting parallels between hierarchical CG models and other deep architectures. %Z Reissued by PMLR on 04 October 2026.
APA
Jojic, N., Perina, A. & Kim, D.. (2016). Hierarchical learning of grids of microtopics. Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R14:148-157 Available from https://proceedings.mlr.press/r14/jojic16a.html. Reissued by PMLR on 04 October 2026.

Related Material