A Caltech Library Service

Active Learning of Multiple Source Multiple Destination Topologies

Sattari, Pegah and Kurant, Maciej and Anandkumar, Animashree and Markopoulou, Athina and Rabbat, Michael G. (2014) Active Learning of Multiple Source Multiple Destination Topologies. IEEE Transactions on Signal Processing, 62 (8). pp. 1926-1937. ISSN 1053-587X. doi:10.1109/TSP.2014.2304431.

[img] PDF - Submitted Version
See Usage Policy.


Use this Persistent URL to link to this item:


We consider the problem of inferring the topology of a network with M sources and N receivers (an M-by- N network), by sending probes between the sources and receivers. Prior work has shown that this problem can be decomposed into two parts: first, infer smaller subnetwork components (1-by- N's or 2-by-2's) and then merge them to identify the M-by- N topology. We focus on the second part, which had previously received less attention in the literature. We assume that a 1-by- N topology is given and that all 2-by-2 components can be queried and learned using end-to-end probes. The problem is which 2-by-2's to query and how to merge them with the given 1-by- N, so as to exactly identify the 2-by- N topology, and optimize a number of performance metrics, including the number of queries (which directly translates into measurement bandwidth), time complexity, and memory usage. We provide a lower bound, [N/2], on the number of 2-by-2's required by any active learning algorithm and propose two greedy algorithms. The first algorithm follows the framework of multiple hypothesis testing, in particular Generalized Binary Search (GBS). The second algorithm is called the Receiver Elimination Algorithm (REA) and follows a bottom-up approach. It requires exactly N-1 steps, which is much less than all (2N) possible 2-by-2's. Simulation results demonstrate that both algorithms correctly identify the 2-by- N topology and are near-optimal, but REA is more efficient in practice.

Item Type:Article
Related URLs:
URLURL TypeDescription ItemConference Paper Paper
Additional Information:© 2014 IEEE. Manuscript received July 27, 2013; accepted January 16, 2014. Date of publication February 04, 2014; date of current version March 17, 2014. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Shuguang (Robert) Cui. This work was supported by an NSF Award 1028394, AFOSR Award FA9550-10-1-0310 and AFOSR MURI FA9550-09-0643. The work of M. Rabbat was supported in part by the Natural Sciences and Engineering Research Council of Canada.
Funding AgencyGrant Number
Air Force Office of Scientific Research (AFOSR)FA9550-10-1-0310
Air Force Office of Scientific Research (AFOSR)FA9550-09-0643
Natural Sciences and Engineering Research Council of Canada (NSERC)UNSPECIFIED
Subject Keywords:Active hypothesis testing, adaptive sensing algorithms, applications of statistical signal processing techniques, inference and estimation on graphs, Internet, network monitoring, sequential learning, tomography
Issue or Number:8
Record Number:CaltechAUTHORS:20170925-101553300
Persistent URL:
Official Citation:P. Sattari, M. Kurant, A. Anandkumar, A. Markopoulou and M. G. Rabbat, "Active Learning of Multiple Source Multiple Destination Topologies," in IEEE Transactions on Signal Processing, vol. 62, no. 8, pp. 1926-1937, April15, 2014. doi: 10.1109/TSP.2014.2304431 URL:
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:81805
Deposited By: Tony Diaz
Deposited On:25 Sep 2017 17:26
Last Modified:15 Nov 2021 19:46

Repository Staff Only: item control page