Structure Learning of Partitioned Markov Networks

Song Liu, Taiji Suzuki, Masashi Sugiyama, Kenji Fukumizu
; Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:439-448, 2016.

Abstract

We learn the structure of a Markov Network between two groups of random variables from joint observations. Since modelling and learning the full MN structure may be hard, learning the links between two groups directly may be a preferable option. We introduce a novel concept called the \emphpartitioned ratio whose factorization directly associates with the Markovian properties of random variables across two groups. A simple one-shot convex optimization procedure is proposed for learning the \emphsparse factorizations of the partitioned ratio and it is theoretically guaranteed to recover the correct inter-group structure under mild conditions. The performance of the proposed method is experimentally compared with the state of the art MN structure learning methods using ROC curves. Real applications on analyzing bipartisanship in US congress and pairwise DNA/time-series alignments are also reported.

Cite this Paper


BibTeX
@InProceedings{pmlr-v48-liuc16, title = {Structure Learning of Partitioned Markov Networks}, author = {Song Liu and Taiji Suzuki and Masashi Sugiyama and Kenji Fukumizu}, booktitle = {Proceedings of The 33rd International Conference on Machine Learning}, pages = {439--448}, year = {2016}, editor = {Maria Florina Balcan and Kilian Q. Weinberger}, volume = {48}, series = {Proceedings of Machine Learning Research}, address = {New York, New York, USA}, month = {20--22 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v48/liuc16.pdf}, url = {http://proceedings.mlr.press/v48/liuc16.html}, abstract = {We learn the structure of a Markov Network between two groups of random variables from joint observations. Since modelling and learning the full MN structure may be hard, learning the links between two groups directly may be a preferable option. We introduce a novel concept called the \emphpartitioned ratio whose factorization directly associates with the Markovian properties of random variables across two groups. A simple one-shot convex optimization procedure is proposed for learning the \emphsparse factorizations of the partitioned ratio and it is theoretically guaranteed to recover the correct inter-group structure under mild conditions. The performance of the proposed method is experimentally compared with the state of the art MN structure learning methods using ROC curves. Real applications on analyzing bipartisanship in US congress and pairwise DNA/time-series alignments are also reported.} }
Endnote
%0 Conference Paper %T Structure Learning of Partitioned Markov Networks %A Song Liu %A Taiji Suzuki %A Masashi Sugiyama %A Kenji Fukumizu %B Proceedings of The 33rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2016 %E Maria Florina Balcan %E Kilian Q. Weinberger %F pmlr-v48-liuc16 %I PMLR %J Proceedings of Machine Learning Research %P 439--448 %U http://proceedings.mlr.press %V 48 %W PMLR %X We learn the structure of a Markov Network between two groups of random variables from joint observations. Since modelling and learning the full MN structure may be hard, learning the links between two groups directly may be a preferable option. We introduce a novel concept called the \emphpartitioned ratio whose factorization directly associates with the Markovian properties of random variables across two groups. A simple one-shot convex optimization procedure is proposed for learning the \emphsparse factorizations of the partitioned ratio and it is theoretically guaranteed to recover the correct inter-group structure under mild conditions. The performance of the proposed method is experimentally compared with the state of the art MN structure learning methods using ROC curves. Real applications on analyzing bipartisanship in US congress and pairwise DNA/time-series alignments are also reported.
RIS
TY - CPAPER TI - Structure Learning of Partitioned Markov Networks AU - Song Liu AU - Taiji Suzuki AU - Masashi Sugiyama AU - Kenji Fukumizu BT - Proceedings of The 33rd International Conference on Machine Learning PY - 2016/06/11 DA - 2016/06/11 ED - Maria Florina Balcan ED - Kilian Q. Weinberger ID - pmlr-v48-liuc16 PB - PMLR SP - 439 DP - PMLR EP - 448 L1 - http://proceedings.mlr.press/v48/liuc16.pdf UR - http://proceedings.mlr.press/v48/liuc16.html AB - We learn the structure of a Markov Network between two groups of random variables from joint observations. Since modelling and learning the full MN structure may be hard, learning the links between two groups directly may be a preferable option. We introduce a novel concept called the \emphpartitioned ratio whose factorization directly associates with the Markovian properties of random variables across two groups. A simple one-shot convex optimization procedure is proposed for learning the \emphsparse factorizations of the partitioned ratio and it is theoretically guaranteed to recover the correct inter-group structure under mild conditions. The performance of the proposed method is experimentally compared with the state of the art MN structure learning methods using ROC curves. Real applications on analyzing bipartisanship in US congress and pairwise DNA/time-series alignments are also reported. ER -
APA
Liu, S., Suzuki, T., Sugiyama, M. & Fukumizu, K.. (2016). Structure Learning of Partitioned Markov Networks. Proceedings of The 33rd International Conference on Machine Learning, in PMLR 48:439-448

Related Material