My current research interests are fundamentally rooted in the areas of machine learning and network analysis. For over a decade, I have developed and studied large margin machine learning algorithms in the context of a range of perception tasks, such as object classification in video, sonar classification, hyperspectral anomaly detection and acoustic classification. In recent years, I have turned my attention to the field of social network analysis and the plethora of challenges associated with understanding dynamic social networks in the online realm.
My core interests in the theoretical aspects of statistical machine learning are driven by the following fundamental questions:
How can machines efficiently acquire context from a human expert over time to solve a pattern recognition task?
How can a machine identify deficiencies in its own understanding of the environment in order to request more context from the user?
How does one estimate the generalization performance of a model efficiently?
What approaches allow a designer to impose explicit constraints on the computational complexity of a classifier during learning?
These questions have motivated my work in incremental and active learning, novelty detection, generalization performance assessment and sparse kernel machine learning.
My pursuits in machine learning are motivated by an overarching desire to build analytic processes that aid a user/analyst in deriving and capturing trusted insights in a timely fashion from large data sources. In any exploratory process, the analyst will want to make specifications of relevance that a machine can learn from to accelerate the process of insight generation. This is inherently a collaborative process between the analyst and machine, requiring a synnergistic mix of machine learning, visualization and interaction paradigms tailored to the analytic task at hand.
Network analysis provides a rich domain in which to develop and apply these concepts. As social media technologies transform the ways in which millions of individuals communicate and collaborate each day, there are new opportunities to observe and understand how networks develop and evolve online. I am particularly interested in methods that help us understand the evolution of social networks within formal organizations and networked groups. As we improve our capabilities to understand factors that are driving the evolution of a network, we may then move toward analytic processes that help decisionmakers take actions to purposefully amplify or degrade the future performance of the network as needed.
"SocialRank: An Ego- and Time-Centric Workflow for Social Relationship Identification," 2008 ARO-USMA Network Science Workshop [PDF]
"Relationship Identification for Social Network Discovery," 2008 AAAI Enhanced Messaging Workshop [PDF]
"Collaborative Social Network Discovery from Online Communications," 2007 ARI-USMA Network Science Workshop [PDF]
"Approximate Leave-One-Out Error Estimation for Learning with Smooth, Strictly Convex Margin Loss Functions," 2004 IEEE Workshop on Machine Learning for Signal Processing [PDF]
"SVM Incremental Learning, Adaptation and Optimization," 2003 International Joint Conference on Neural Networks [PDF]
J. Montemayor, C. P. Diehl, M. Pekala, D. Patrone, "SocialRank: An Ego- and Time-Centric Workflow for Relationship Identification," 2008 IEEE Symposium on Visual Analytics Science and Technology [PDF]
C. P. Diehl, A. Llorens, "Data-Dependent Generalization Performance Assessment Via Quasiconvex Optimization," 2008 IEEE International Workshop on Machine Learning for Signal Processing [PDF]
C. P. Diehl, G. Namata, L. Getoor, "Relationship Identification for Social Network Discovery," AAAI 2008 Enhanced Messaging Workshop [PDF]
(Note: this paper is a shortened version of the AAAI 2007 paper)
C. P. Diehl, G. Namata, L. Getoor, "Relationship Identification for Social Network Discovery," AAAI 2007 [PDF]
A. Banerjee, P. Burlina, C. P. Diehl, "A Support Vector Method for Anomaly Detection in Hyperspectral Imagery,"
IEEE Transactions on Geoscience and Remote Sensing, 44(8), August 2006, pp. 2282-2291 [PDF]
C. P. Diehl, L. Getoor, G. Namata, "Name Reference Resolution in Organizational Email Archives," 2006 SIAM International Conference on Data Mining [PDF]
L. Getoor, C. P. Diehl, "Link Mining: A Survey," ACM SIGKDD Explorations, 7(2), December 2005 [PDF]
L. Getoor, C. P. Diehl eds., ACM SIGKDD Explorations Special Issue on Link Mining, 7(2), December 2005
G. Cauwenberghs, A. Andreou, J. West, M. Stanacevic, A. Celik, P. Julian, T. Teixeira, C. Diehl, L. Riddle, "A Miniature, Low-Power, Intelligent Sensor Node for Persistent Acoustic Surveillance," Proceedings of the SPIE, 5796, May 2005, pp. 294-305 [PDF]
C. J. Costello, C. P. Diehl, A. Banerjee, H. Fisher, "Scheduling an Active Camera to Observe People," Proceedings of the 2nd ACM International Workshop on Video Surveillance and Sensor Networks, October 2004 [PDF]
R. J. Vogelstein, K. Murari, P. H. Thakur, G. Cauwenberghs, S. Chakrabartty, C. P. Diehl, "Spike Sorting with Support Vector Machines," Proceedings of the 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, San Francisco, CA, September 2004 [PDF]
C. P. Diehl, "Approximate Leave-One-Out Error Estimation for Learning with Smooth, Strictly Convex Margin Loss Functions," Proceedings of the 2004 IEEE Workshop on Machine Learning for Signal Processing [PDF]
I-J. Wang, C. P. Diehl, F. J. Pineda, "A Statistical Model of Proteolytic Digestion," Proceedings of the 2003 IEEE Computer Society Bioinformatics Conference, pp. 506-508 [PDF]
C. P. Diehl, G. Cauwenberghs, "SVM Incremental Learning, Adaptation and Optimization," Proceedings of the 2003 International Joint Conference on Neural Networks, Special Session on Incremental Learning, pp. 2685-2690, Invited Paper [PDF]
M. Saptharishi, C. S. Oliver, C. P. Diehl, K. S. Bhat, J. M. Dolan, A. Trebi-Ollennu, P. K. Khosla, "Distributed Surveillance and Reconnaissance Using Multiple Autonomous ATVs: CyberScout," IEEE Transactions on Robotics and Automation: Special Issue on Advances in Multirobot Systems, vol. 18, no. 5, October 2002 [PDF]
C. P. Diehl, J. B. Hampshire II, "Real-time Object Classification and Novelty Detection for Collaborative Video Surveillance,"Proceedings of the 2002 International Joint Conference on Neural Networks, vol. 3, pp. 2620-2625 [PDF]
C. P. Diehl, "Toward Efficient Collaborative Classification for Distributed Video Surveillance," Ph.D. Thesis, Department of Electrical and Computer Engineering, Carnegie Mellon University, December 2000
M. Saptharishi, K. S. Bhat, C. P. Diehl, J. M. Dolan, P. K. Khosla, "CyberScout: Distributed Agents for Autonomous Reconnaissance and Surveillance," Proceedings of the Conference on Mechatronics and Machine Vision in Practice, September 2000, pp. 93-100 [PDF]
M. Saptharishi, K. Bhat, C. Diehl, C. Oliver, M. Savvides, A. Soto, J. Dolan, P. Khosla, "Recent Advances in Distributed Collaborative, Surveillance," Proceedings of the SPIE, vol. 4040, pp. 199-208 [PDF]
C. P. Diehl, M. Saptharishi, J. B. Hampshire II, P. K. Khosla, "Collaborative Surveillance Using Both Fixed and Mobile Unattended Ground Sensors," Proceedings of the SPIE, vol. 3713, July 1999, pp. 178-185 [PDF]
C. P. Diehl, B. E. Henty, N. Kanodia, D. D. Stancil, "Wireless RF Distribution in Buildings Using Heating and Ventilation Ducts," Proceedings of 8th Annual Virginia Tech Symposium on Wireless Personal Communications, Blacksburg, VA, 10-12 June 1998, pp. 61-70 [PDF]
C. P. Diehl, "Rademacher Penalty Optimization with the Generalized Ramp Loss", 2008. [PDF]
Chris.Diehl at jhuapl dot edu
443.778.3457 (office)
443.778.6904 (fax)
Dr. Chris Diehl
Johns Hopkins Applied Physics Laboratory
11100 Johns Hopkins Road
Laurel, Maryland 20723