We consider the problems of detection and support recovery of a contiguous block of weak activation in a large matrix, from noisy, possibly adaptively chosen, compressive (linear) measurements. We precisely characterize the tradeoffs between the various problem dimensions, the signal strength and the number of measurements required to reliably detect and recover the support of the signal, both for passive and adaptive measurement schemes. In each case, we complement algorithmic results with information-theoretic lower bounds. Analogous to the situation in the closely related problem of noisy compressed sensing, we show that for detection neither adaptivity, nor structure reduce the minimax signal strength requirement. On the other hand we show the rather surprising result that, contrary to the situation in noisy compressed sensing, the signal strength requirement to recover the support of a contiguous block-structured signal is strongly influenced by both the signal structure and the ability to choose measurements adaptively.
References
[1] S. Aeron, V. Saligrama, and M. Zhao. Information theoretic bounds for compressed sensing., IEEE Trans. Inf. Theory, 56(10) :5111–5130, 2010. 1366.94179 10.1109/TIT.2010.2059891[1] S. Aeron, V. Saligrama, and M. Zhao. Information theoretic bounds for compressed sensing., IEEE Trans. Inf. Theory, 56(10) :5111–5130, 2010. 1366.94179 10.1109/TIT.2010.2059891
[2] E. Arias-Castro. Detecting a vector based on linear measurements., Electron. J. Stat., 6:547–558, 2012. 1274.62378 10.1214/12-EJS686[2] E. Arias-Castro. Detecting a vector based on linear measurements., Electron. J. Stat., 6:547–558, 2012. 1274.62378 10.1214/12-EJS686
[3] E. Arias-Castro, E. J. Candés, and M. A. Davenport. On the fundamental limits of adaptive sensing., IEEE Trans. Inf. Theory, 59(1):472–481, 2013. 1364.94106 10.1109/TIT.2012.2215837[3] E. Arias-Castro, E. J. Candés, and M. A. Davenport. On the fundamental limits of adaptive sensing., IEEE Trans. Inf. Theory, 59(1):472–481, 2013. 1364.94106 10.1109/TIT.2012.2215837
[4] E. Arias-Castro, E. J. Candés, and A. Durand. Detection of an anomalous cluster in a network., Ann. Stat., 39(1):278–304, 2011a. 1209.62097 10.1214/10-AOS839 euclid.aos/1291388376[4] E. Arias-Castro, E. J. Candés, and A. Durand. Detection of an anomalous cluster in a network., Ann. Stat., 39(1):278–304, 2011a. 1209.62097 10.1214/10-AOS839 euclid.aos/1291388376
[5] E. Arias-Castro, E. J. Candés, and Y. Plan. Global testing under sparse alternatives: Anova, multiple comparisons and the higher criticism., Ann. Stat., 39(5) :2533–2556, 2011b. 1231.62136 10.1214/11-AOS910 euclid.aos/1322663467[5] E. Arias-Castro, E. J. Candés, and Y. Plan. Global testing under sparse alternatives: Anova, multiple comparisons and the higher criticism., Ann. Stat., 39(5) :2533–2556, 2011b. 1231.62136 10.1214/11-AOS910 euclid.aos/1322663467
[6] S. Balakrishnan, M. Kolar, A. Rinaldo, and A. Singh. Recovering block-structured activations using compressive measurements. 2012., arXiv:1209.3431. 1366.62034 10.1214/17-EJS1267[6] S. Balakrishnan, M. Kolar, A. Rinaldo, and A. Singh. Recovering block-structured activations using compressive measurements. 2012., arXiv:1209.3431. 1366.62034 10.1214/17-EJS1267
[7] R. G. Baraniuk, V. Cevher, M. F. Duarte, and C. Hegde. Model-based compressive sensing., IEEE Trans. Inf. Theory, 56(4) :1982–2001, 2010. 1366.94215 10.1109/TIT.2010.2040894[7] R. G. Baraniuk, V. Cevher, M. F. Duarte, and C. Hegde. Model-based compressive sensing., IEEE Trans. Inf. Theory, 56(4) :1982–2001, 2010. 1366.94215 10.1109/TIT.2010.2040894
[8] S. Bhamidi, P. S. Dey, and A. B. Nobel. Energy landscape for large average submatrix detection problems in gaussian random matrices. 2012., arXiv:1211.2284. 1371.60010 10.1007/s00440-017-0766-0[8] S. Bhamidi, P. S. Dey, and A. B. Nobel. Energy landscape for large average submatrix detection problems in gaussian random matrices. 2012., arXiv:1211.2284. 1371.60010 10.1007/s00440-017-0766-0
[9] C. Butucea and Y. I. Ingster. Detection of a sparse submatrix of a high-dimensional noisy matrix., Bernoulli, 19(5B) :2652–2688, 2013. 06254575 10.3150/12-BEJ470 euclid.bj/1386078616[9] C. Butucea and Y. I. Ingster. Detection of a sparse submatrix of a high-dimensional noisy matrix., Bernoulli, 19(5B) :2652–2688, 2013. 06254575 10.3150/12-BEJ470 euclid.bj/1386078616
[10] C. Butucea, Y. I. Ingster, and I. Suslina. Sharp variable selection of a sparse submatrix in a high-dimensional noisy matrix. 2013., arXiv:1303.5647. 1330.62169 10.1051/ps/2014017[10] C. Butucea, Y. I. Ingster, and I. Suslina. Sharp variable selection of a sparse submatrix in a high-dimensional noisy matrix. 2013., arXiv:1303.5647. 1330.62169 10.1051/ps/2014017
[11] E. J. Candés and M. A. Davenport. How well can we estimate a sparse vector?, Appl. Comput. Harmon. Anal., 34(2):317–323, 2013. 1315.94019 10.1016/j.acha.2012.08.010[11] E. J. Candés and M. A. Davenport. How well can we estimate a sparse vector?, Appl. Comput. Harmon. Anal., 34(2):317–323, 2013. 1315.94019 10.1016/j.acha.2012.08.010
[12] E. J. Candés and T. Tao. The Dantzig selector: Statistical estimation when $p$ is much larger than $n$., Ann. Stat., 35(6) :2313–2351, 2007. 1139.62019 10.1214/009053606000001523 euclid.aos/1201012958[12] E. J. Candés and T. Tao. The Dantzig selector: Statistical estimation when $p$ is much larger than $n$., Ann. Stat., 35(6) :2313–2351, 2007. 1139.62019 10.1214/009053606000001523 euclid.aos/1201012958
[14] Y. Caron, P. Makris, and N. Vincent. A method for detecting artificial objects in natural environments. In, 16th Int. Conf. Pattern Recogn., volume 1, pages 600–603. IEEE, 2002.[14] Y. Caron, P. Makris, and N. Vincent. A method for detecting artificial objects in natural environments. In, 16th Int. Conf. Pattern Recogn., volume 1, pages 600–603. IEEE, 2002.
[15] R. M. Castro. Adaptive sensing performance lower bounds for sparse signal detection and support estimation. 2012., arXiv:1206.0648.[15] R. M. Castro. Adaptive sensing performance lower bounds for sparse signal detection and support estimation. 2012., arXiv:1206.0648.
[16] R. M. Castro and E. Tánczos. Adaptive compressed sensing for estimation of structured sparse sets. 2014., arXiv:1410.4593.[16] R. M. Castro and E. Tánczos. Adaptive compressed sensing for estimation of structured sparse sets. 2014., arXiv:1410.4593.
[17] O. Catoni. Challenging the empirical mean and empirical variance: a deviation study., Ann. Inst. Henri Poincaré Probab. Stat., 48(4) :1148–1185, 2012. 1282.62070 10.1214/11-AIHP454 euclid.aihp/1353098444[17] O. Catoni. Challenging the empirical mean and empirical variance: a deviation study., Ann. Inst. Henri Poincaré Probab. Stat., 48(4) :1148–1185, 2012. 1282.62070 10.1214/11-AIHP454 euclid.aihp/1353098444
[18] S. Chatterjee. Matrix estimation by universal singular value thresholding., Ann. Statist., 43(1):177–214, 2015. 1308.62038 10.1214/14-AOS1272 euclid.aos/1418135619[18] S. Chatterjee. Matrix estimation by universal singular value thresholding., Ann. Statist., 43(1):177–214, 2015. 1308.62038 10.1214/14-AOS1272 euclid.aos/1418135619
[19] M. A. Davenport and E. Arias-Castro. Compressive binary search. 2012., arXiv:1202.0937.[19] M. A. Davenport and E. Arias-Castro. Compressive binary search. 2012., arXiv:1202.0937.
[20] D. L. Donoho. Compressed sensing., IEEE Trans. Inf. Theory, 52(4) :1289–1306, 2006. 1288.94016 10.1109/TIT.2006.871582[20] D. L. Donoho. Compressed sensing., IEEE Trans. Inf. Theory, 52(4) :1289–1306, 2006. 1288.94016 10.1109/TIT.2006.871582
[21] M. F. Duarte, M. A. Davenport, M. J. Wainwright, and R. G. Baraniuk. Sparse signal detection from incoherent projections. In, Proc. IEEE Int. Conf. Acoustics Speed and Signal Processing, pages III–305–III–308. 2006.[21] M. F. Duarte, M. A. Davenport, M. J. Wainwright, and R. G. Baraniuk. Sparse signal detection from incoherent projections. In, Proc. IEEE Int. Conf. Acoustics Speed and Signal Processing, pages III–305–III–308. 2006.
[22] C. F. F. C. Filho, R. de Oliveira Melo, and M. G. F. Costa., Detecting Natural Gas Leaks Using Digital Images and Novelty Filters, pages 242–249. Springer Berlin Heidelberg, Berlin, Heidelberg, 2012.[22] C. F. F. C. Filho, R. de Oliveira Melo, and M. G. F. Costa., Detecting Natural Gas Leaks Using Digital Images and Novelty Filters, pages 242–249. Springer Berlin Heidelberg, Berlin, Heidelberg, 2012.
[23] G. M. Foody and A. Mathur. A relative evaluation of multiclass image classification by support vector machines., IEEE Transactions on Geoscience and Remote Sensing, 42(6) :1335–1343, 2004.[23] G. M. Foody and A. Mathur. A relative evaluation of multiclass image classification by support vector machines., IEEE Transactions on Geoscience and Remote Sensing, 42(6) :1335–1343, 2004.
[24] M. Golbabaee and P. Vandergheynst. Compressed sensing of simultaneous low-rank and joint-sparse matrices. 2012., arXiv:1211.5058.[24] M. Golbabaee and P. Vandergheynst. Compressed sensing of simultaneous low-rank and joint-sparse matrices. 2012., arXiv:1211.5058.
[25] J. D. Haupt, R. G. Baraniuk, R. M. Castro, and R. D. Nowak. Compressive distilled sensing: Sparse recovery using adaptivity in compressive measurements. In, Proc. 43rd Asilomar Conf. Signals, Systems and Computers, pages 1551–1555. IEEE, 2009.[25] J. D. Haupt, R. G. Baraniuk, R. M. Castro, and R. D. Nowak. Compressive distilled sensing: Sparse recovery using adaptivity in compressive measurements. In, Proc. 43rd Asilomar Conf. Signals, Systems and Computers, pages 1551–1555. IEEE, 2009.
[26] J. D. Haupt and R. D. Nowak. Compressive sampling for signal detection. In, Proc. IEEE Int. Conf. Acoustics, Speech and Signal Processing, volume 3, pages III –1509–III–1512. IEEE, 2007.[26] J. D. Haupt and R. D. Nowak. Compressive sampling for signal detection. In, Proc. IEEE Int. Conf. Acoustics, Speech and Signal Processing, volume 3, pages III –1509–III–1512. IEEE, 2007.
[27] J. Huang, T. Zhang, and D. Metaxas. Learning with structured sparsity., J. Mach. Learn. Res., 12 :3371–3412, 2011. 1280.68169[27] J. Huang, T. Zhang, and D. Metaxas. Learning with structured sparsity., J. Mach. Learn. Res., 12 :3371–3412, 2011. 1280.68169
[28] Y. I. Ingster, A. B. Tsybakov, and N. Verzelen. Detection boundary in sparse regression., Electron. J. Stat., 4 :1476–1526, 2010. 1329.62314 10.1214/10-EJS589[28] Y. I. Ingster, A. B. Tsybakov, and N. Verzelen. Detection boundary in sparse regression., Electron. J. Stat., 4 :1476–1526, 2010. 1329.62314 10.1214/10-EJS589
[29] M. Kolar, S. Balakrishnan, A. Rinaldo, and A. Singh. Minimax localization of structural information in large noisy matrices. In J. Shawe-Taylor, R. S. Zemel, P. L. Bartlett, F. C. N. Pereira, and K. Q. Weinberger, editors, Advances in Neural Information Processing Systems 24, pages 909–917. 2011.[29] M. Kolar, S. Balakrishnan, A. Rinaldo, and A. Singh. Minimax localization of structural information in large noisy matrices. In J. Shawe-Taylor, R. S. Zemel, P. L. Bartlett, F. C. N. Pereira, and K. Q. Weinberger, editors, Advances in Neural Information Processing Systems 24, pages 909–917. 2011.
[30] V. Koltchinskii, K. Lounici, and A. B. Tsybakov. Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion., Ann. Stat., 39(5) :2302–2329, 2011. 1231.62097 10.1214/11-AOS894 euclid.aos/1322663459[30] V. Koltchinskii, K. Lounici, and A. B. Tsybakov. Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion., Ann. Stat., 39(5) :2302–2329, 2011. 1231.62097 10.1214/11-AOS894 euclid.aos/1322663459
[31] A. Krishnamurthy, J. Sharpnack, and A. Singh. Recovering graph-structured activations using adaptive compressive measurements. 2013., arXiv:1305.0213.[31] A. Krishnamurthy, J. Sharpnack, and A. Singh. Recovering graph-structured activations using adaptive compressive measurements. 2013., arXiv:1305.0213.
[32] Y. Ma, D. J. Sutherland, R. Garnett, and J. G. Schneider. Active pointillistic pattern search. In, AISTATS, volume 38 of JMLR Workshop and Conference Proceedings. JMLR.org, 2015.[32] Y. Ma, D. J. Sutherland, R. Garnett, and J. G. Schneider. Active pointillistic pattern search. In, AISTATS, volume 38 of JMLR Workshop and Conference Proceedings. JMLR.org, 2015.
[33] M. L. Malloy and R. D. Nowak. Near-optimal adaptive compressed sensing. In, Proc. 46th Asilomar Conf. Signals, Systems and Computers, pages 1935–1939. IEEE, 2012a.[33] M. L. Malloy and R. D. Nowak. Near-optimal adaptive compressed sensing. In, Proc. 46th Asilomar Conf. Signals, Systems and Computers, pages 1935–1939. IEEE, 2012a.
[34] M. L. Malloy and R. D. Nowak. Near-optimal compressive binary search. 2012b., arXiv:1203.1804.[34] M. L. Malloy and R. D. Nowak. Near-optimal compressive binary search. 2012b., arXiv:1203.1804.
[35] S. Negahban and M. J. Wainwright. Estimation of (near) low-rank matrices with noise and high-dimensional scaling., Ann. Stat., 39(2) :1069–1097, 2011. 1216.62090 10.1214/10-AOS850 euclid.aos/1304947044[35] S. Negahban and M. J. Wainwright. Estimation of (near) low-rank matrices with noise and high-dimensional scaling., Ann. Stat., 39(2) :1069–1097, 2011. 1216.62090 10.1214/10-AOS850 euclid.aos/1304947044
[36] G. Reeves and M. Gastpar. The sampling rate-distortion tradeoff for sparsity pattern recovery in compressed sensing., IEEE Trans. Inf. Theory, 58(5) :3065–3092, 2012. 1365.94184 10.1109/TIT.2012.2184848[36] G. Reeves and M. Gastpar. The sampling rate-distortion tradeoff for sparsity pattern recovery in compressed sensing., IEEE Trans. Inf. Theory, 58(5) :3065–3092, 2012. 1365.94184 10.1109/TIT.2012.2184848
[37] E. Richard, P.-A. Savalle, and N. Vayatis. Estimation of simultaneously sparse and low rank matrices. In, Proc. 29th Int. Conf. Mach. Learn., pages 1351–1358. 2012.[37] E. Richard, P.-A. Savalle, and N. Vayatis. Estimation of simultaneously sparse and low rank matrices. In, Proc. 29th Int. Conf. Mach. Learn., pages 1351–1358. 2012.
[38] A. Soni and J. D. Haupt. Efficient adaptive compressive sensing using sparse hierarchical learned dictionaries. In, Proc. 45th Conf. Signals, Systems and Computers, pages 1250–1254. IEEE, 2011.[38] A. Soni and J. D. Haupt. Efficient adaptive compressive sensing using sparse hierarchical learned dictionaries. In, Proc. 45th Conf. Signals, Systems and Computers, pages 1250–1254. IEEE, 2011.
[39] A. Soni and J. D. Haupt. On the fundamental limits of recovering tree sparse vectors from noisy linear measurements. 2013., arXiv:1306.4391. 1364.94159 10.1109/TIT.2013.2287496[39] A. Soni and J. D. Haupt. On the fundamental limits of recovering tree sparse vectors from noisy linear measurements. 2013., arXiv:1306.4391. 1364.94159 10.1109/TIT.2013.2287496
[40] X. Sun and A. B. Nobel. On the maximal size of large-average and ANOVA-fit submatrices in a gaussian random matrix., Bernoulli, 19(1):275–294, 2013. 1259.62062 10.3150/11-BEJ394 euclid.bj/1358531750[40] X. Sun and A. B. Nobel. On the maximal size of large-average and ANOVA-fit submatrices in a gaussian random matrix., Bernoulli, 19(1):275–294, 2013. 1259.62062 10.3150/11-BEJ394 euclid.bj/1358531750
[41] E. Tánczos and R. M. Castro. Adaptive sensing for estimation of structured sparse signals. 2013., arXiv:1311.7118.[41] E. Tánczos and R. M. Castro. Adaptive sensing for estimation of structured sparse signals. 2013., arXiv:1311.7118.
[43] D. Tuia, F. Ratle, F. Pacifici, M. F. Kanevski, and W. J. Emery. Active learning methods for remote sensing image classification., IEEE Transactions on Geoscience and Remote Sensing, 47(7) :2218–2232, 2009.[43] D. Tuia, F. Ratle, F. Pacifici, M. F. Kanevski, and W. J. Emery. Active learning methods for remote sensing image classification., IEEE Transactions on Geoscience and Remote Sensing, 47(7) :2218–2232, 2009.
[44] M. M. Wagner, F. C. Tsui, J. U. Espino, V. M. Dato, D. F. Sittig, R. A. Caruana, L. F. McGinnis, D. W. Deerfield, M. J. Druzdzel, and D. B. Fridsma. The emerging science of very early detection of disease outbreaks., J Public Health Manag Pract, 7(6):51–59, 2001.[44] M. M. Wagner, F. C. Tsui, J. U. Espino, V. M. Dato, D. F. Sittig, R. A. Caruana, L. F. McGinnis, D. W. Deerfield, M. J. Druzdzel, and D. B. Fridsma. The emerging science of very early detection of disease outbreaks., J Public Health Manag Pract, 7(6):51–59, 2001.
[45] M. J. Wainwright. Information-theoretic limits on sparsity recovery in the high-dimensional and noisy setting., IEEE Trans. Inf. Theory, 55(12) :5728–5741, 2009a. 1367.94106 10.1109/TIT.2009.2032816[45] M. J. Wainwright. Information-theoretic limits on sparsity recovery in the high-dimensional and noisy setting., IEEE Trans. Inf. Theory, 55(12) :5728–5741, 2009a. 1367.94106 10.1109/TIT.2009.2032816
[46] M. J. Wainwright. Sharp thresholds for high-dimensional and noisy sparsity recovery using $\ell_1$-constrained quadratic programming (lasso)., IEEE Trans. Inf. Theory, 55(5) :2183–2202, 2009b. 1367.62220 10.1109/TIT.2009.2016018[46] M. J. Wainwright. Sharp thresholds for high-dimensional and noisy sparsity recovery using $\ell_1$-constrained quadratic programming (lasso)., IEEE Trans. Inf. Theory, 55(5) :2183–2202, 2009b. 1367.62220 10.1109/TIT.2009.2016018
[47] S. Yoon, C. Nardini, L. Benini, and G. De Micheli. Discovering coherent biclusters from gene expression data using zero-suppressed binary decision diagrams., IEEE/ACM Trans. Comput. Biol. Bioinf., 2(4):339–354, 2005.[47] S. Yoon, C. Nardini, L. Benini, and G. De Micheli. Discovering coherent biclusters from gene expression data using zero-suppressed binary decision diagrams., IEEE/ACM Trans. Comput. Biol. Bioinf., 2(4):339–354, 2005.
