Approximate Inference in Additive Factorial HMMs with Application to Energy Disaggregation

J. Zico Kolter, Tommi Jaakkola
Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, PMLR 22:1472-1482, 2012.

Abstract

This paper considers additive factorial hidden Markov models, an extension to HMMs where the state factors into multiple independent chains, and the output is an additive function of all the hidden states. Although such models are very powerful, accurate inference is unfortunately difficult: exact inference is not computationally tractable, and existing approximate inference techniques are highly susceptible to local optima. In this paper we propose an alternative inference method for such models, which exploits their additive structure by 1) looking at the observed difference signal of the observation, 2) incorporating a “robust” mixture component that can account for unmodeled observations, and 3) constraining the posterior to allow at most one hidden state to change at a time. Combining these elements we develop a convex formulation of approximate inference that is computationally efficient, has no issues of local optima, and which performs much better than existing approaches in practice. The method is motivated by the problem of energy disaggregation, the task of taking a whole home electricity signal and decomposing it into its component appliances; applied to this task, our algorithm achieves state-of-the-art performance, and is able to separate many appliances almost perfectly using just the total aggregate signal.

Cite this Paper

BibTeX
@InProceedings{pmlr-v22-zico12, title = {Approximate Inference in Additive Factorial HMMs with Application to Energy Disaggregation}, author = {Kolter, J. Zico and Jaakkola, Tommi}, booktitle = {Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics}, pages = {1472--1482}, year = {2012}, editor = {Lawrence, Neil D. and Girolami, Mark}, volume = {22}, series = {Proceedings of Machine Learning Research}, address = {La Palma, Canary Islands}, month = {21--23 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v22/zico12/zico12.pdf}, url = {https://proceedings.mlr.press/v22/zico12.html}, abstract = {This paper considers additive factorial hidden Markov models, an extension to HMMs where the state factors into multiple independent chains, and the output is an additive function of all the hidden states. Although such models are very powerful, accurate inference is unfortunately difficult: exact inference is not computationally tractable, and existing approximate inference techniques are highly susceptible to local optima. In this paper we propose an alternative inference method for such models, which exploits their additive structure by 1) looking at the observed difference signal of the observation, 2) incorporating a “robust” mixture component that can account for unmodeled observations, and 3) constraining the posterior to allow at most one hidden state to change at a time. Combining these elements we develop a convex formulation of approximate inference that is computationally efficient, has no issues of local optima, and which performs much better than existing approaches in practice. The method is motivated by the problem of energy disaggregation, the task of taking a whole home electricity signal and decomposing it into its component appliances; applied to this task, our algorithm achieves state-of-the-art performance, and is able to separate many appliances almost perfectly using just the total aggregate signal.} }
Endnote
%0 Conference Paper %T Approximate Inference in Additive Factorial HMMs with Application to Energy Disaggregation %A J. Zico Kolter %A Tommi Jaakkola %B Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2012 %E Neil D. Lawrence %E Mark Girolami %F pmlr-v22-zico12 %I PMLR %P 1472--1482 %U https://proceedings.mlr.press/v22/zico12.html %V 22 %X This paper considers additive factorial hidden Markov models, an extension to HMMs where the state factors into multiple independent chains, and the output is an additive function of all the hidden states. Although such models are very powerful, accurate inference is unfortunately difficult: exact inference is not computationally tractable, and existing approximate inference techniques are highly susceptible to local optima. In this paper we propose an alternative inference method for such models, which exploits their additive structure by 1) looking at the observed difference signal of the observation, 2) incorporating a “robust” mixture component that can account for unmodeled observations, and 3) constraining the posterior to allow at most one hidden state to change at a time. Combining these elements we develop a convex formulation of approximate inference that is computationally efficient, has no issues of local optima, and which performs much better than existing approaches in practice. The method is motivated by the problem of energy disaggregation, the task of taking a whole home electricity signal and decomposing it into its component appliances; applied to this task, our algorithm achieves state-of-the-art performance, and is able to separate many appliances almost perfectly using just the total aggregate signal.
RIS
TY - CPAPER TI - Approximate Inference in Additive Factorial HMMs with Application to Energy Disaggregation AU - J. Zico Kolter AU - Tommi Jaakkola BT - Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics DA - 2012/03/21 ED - Neil D. Lawrence ED - Mark Girolami ID - pmlr-v22-zico12 PB - PMLR DP - Proceedings of Machine Learning Research VL - 22 SP - 1472 EP - 1482 L1 - http://proceedings.mlr.press/v22/zico12/zico12.pdf UR - https://proceedings.mlr.press/v22/zico12.html AB - This paper considers additive factorial hidden Markov models, an extension to HMMs where the state factors into multiple independent chains, and the output is an additive function of all the hidden states. Although such models are very powerful, accurate inference is unfortunately difficult: exact inference is not computationally tractable, and existing approximate inference techniques are highly susceptible to local optima. In this paper we propose an alternative inference method for such models, which exploits their additive structure by 1) looking at the observed difference signal of the observation, 2) incorporating a “robust” mixture component that can account for unmodeled observations, and 3) constraining the posterior to allow at most one hidden state to change at a time. Combining these elements we develop a convex formulation of approximate inference that is computationally efficient, has no issues of local optima, and which performs much better than existing approaches in practice. The method is motivated by the problem of energy disaggregation, the task of taking a whole home electricity signal and decomposing it into its component appliances; applied to this task, our algorithm achieves state-of-the-art performance, and is able to separate many appliances almost perfectly using just the total aggregate signal. ER -
APA
Kolter, J.Z. & Jaakkola, T.. (2012). Approximate Inference in Additive Factorial HMMs with Application to Energy Disaggregation. Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 22:1472-1482 Available from https://proceedings.mlr.press/v22/zico12.html.

Related Material