Contextually Supervised Source Separation with Application to Energy Disaggregation

We propose a new framework for single-channel source separation that lies between the fully supervised and unsupervised setting. Instead of supervision, we provide input features for each source signal and use convex methods to estimate the correlations between these features and the unobserved signal decomposition. Contextually supervised source separation is a natural fit for domains with large amounts of data but no explicit supervision; our motivating application is energy disaggregation of hourly smart meter data (the separation of whole-home power signals into different energy uses). Here contextual supervision allows us to provide itemized energy usage for thousands homes, a task previously impossible due to the need for specialized data collection hardware. On smaller datasets which include labels, we demonstrate that contextual supervision improves significantly over a reasonable baseline and existing unsupervised methods for source separation. Finally, we analyze the case of l2 loss theoretically and show that recovery of the signal components depends only on cross-correlation between features for different signals, not on correlations between features for the same signal.

[1]  G. W. Hart,et al.  Nonintrusive appliance load monitoring , 1992, Proc. IEEE.

[2]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..

[3]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[4]  David J. Field,et al.  Sparse coding with an overcomplete basis set: A strategy employed by V1? , 1997, Vision Research.

[5]  Terrence J. Sejnowski,et al.  Learning Overcomplete Representations , 2000, Neural Computation.

[6]  Sam T. Roweis,et al.  One Microphone Source Separation , 2000, NIPS.

[7]  Patrik O. Hoyer,et al.  Non-negative sparse coding , 2002, Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing.

[8]  Michael I. Jordan,et al.  Factorial Hidden Markov Models , 1995, Machine Learning.

[9]  Matthew E. P. Davies,et al.  Shift-invariant sparse coding for single channel blind source separation , 2005 .

[10]  Morten Mørup,et al.  Nonnegative Matrix Factor 2-D Deconvolution for Blind Single Channel Source Separation , 2006, ICA.

[11]  Mikkel N. Schmidt,et al.  Single-channel speech separation using sparse non-negative matrix factorization , 2006, INTERSPEECH.

[12]  Bhiksha Raj,et al.  A Probabilistic Latent Variable Model for Acoustic Modeling , 2006 .

[13]  M.E. Davies,et al.  Source separation using single channel ICA , 2007, Signal Process..

[14]  Andrew Y. Ng,et al.  Energy Disaggregation via Discriminative Sparse Coding , 2010, NIPS.

[15]  Manish Marwah,et al.  Unsupervised Disaggregation of Low Frequency Power Measurements , 2011, SDM.

[16]  Michael Zeifman,et al.  Nonintrusive appliance load monitoring: Review and outlook , 2011, IEEE Transactions on Consumer Electronics.

[17]  Tommi S. Jaakkola,et al.  Approximate Inference in Additive Factorial HMMs with Application to Energy Disaggregation , 2012, AISTATS.

[18]  Alex Rogers,et al.  Non-Intrusive Load Monitoring Using Prior Models of General Appliance Types , 2012, AAAI.