Dr. Georgiy M. Levchuk Profile

Dr. Georgiy M. Levchuk

Senior Principal Research Scientist at Aptima Inc

SPIE Involvement:

Conference Program Committee | Author

Publications (28)

Proceedings Article | 12 April 2021 Presentation + Paper

Learning intent and behavior models from motion trajectories for unsupervised semantic labeling

Georgiy Levchuk, Louis Penafiel

Proceedings Volume 11756, 117560O (2021) https://doi.org/10.1117/12.2589246

Read Abstract +

Proceedings Article | 12 April 2021 Presentation + Paper

Towards adaptive and curious artificial agents: learning, perception, and planning in dynamic uncertain environments

Georgiy Levchuk, Lisa Lucia, Louis Penafiel

Proceedings Volume 11751, 117510D (2021) https://doi.org/10.1117/12.2589259

Read Abstract +

Proceedings Article | 14 May 2019 Presentation

Adaptive data collection and retention planning to support cyber analytics (Conference Presentation)

Georgiy Levchuk

Proceedings Volume 11013, 1101304 (2019) https://doi.org/10.1117/12.2519104

KEYWORDS: Analytics, Computer security, Network security, Sensors, Data modeling, Systems modeling, Data archive systems, Forensic science, Data storage

Read Abstract +

Proceedings Article | 14 May 2019 Presentation

Variational adversarial generative models for information fusion with applications to cross-domain causal reasoning, hypothesis construction and identity learning (Conference Presentation)

Georgiy Levchuk

Proceedings Volume 11018, 110180U (2019) https://doi.org/10.1117/12.2522075

KEYWORDS: Data modeling, Information fusion, Data fusion, Sensors, Signal processing, Gallium nitride, Image fusion, Video, Forensic science, Cognitive modeling

Read Abstract +

Many problems in data fusion require reasoning about heterogeneous data obtained from different sources (news media, first-hand reporters, etc.) and containing diverse modalities (image, video, text, audio, etc.). Construction of the situation hypotheses, predictions of future events or inference of event causes, identification of the actors, or forensic analyses require automated decision aids that can provide fast and reliable extraction of relevant information and association of the sensory inputs. While humans can effectively fuse multiple sensory signals, the machine learning-based methods cannot yet effectively replicate these cognitive processes. The challenges are due to the presence of heterogeneity gap in multi-modal multi-source data, which creates inconsistent distributions and representations of the input signals. Researchers attempted to solve this problem by modeling explicit cross-modal correlation and learning common joint representations of the heterogeneous data. However, only marginal successes have been achieved, and domains have so far been limited to imagery/video-to-text association with no promise to generalize to any knowledge fusion. In this paper, we explore the application of three recent modeling techniques to multi-source multi-modal data fusion: adversarial generative modeling, variational inference, and inverse reinforcement learning. The generative adversarial networks (GANs) are capable of estimating a generative model by an adversarial training process, where the component model that learns generative data distribution competes against the discriminative component that attempts to classify the generated data. GANs-learned distributions are capable of producing the data samples of high quality, which is essential for producing meaningful cross-modality association decisions. First, we use GANs for modeling the joint distribution over the heterogeneous data of different sources and modalities, learning the common representation and improving the cross-modal correlation estimates. Our approach is based on generative modeling of explicit intermodality correlation and intra-modality reconstruction, while the discriminative component will judge the quality of generated samples both within the same modality and between different modalities. Second, we will use the variational inference to add the classification reasoning to our model, making our solution capable of producing both high-quality cross-modal correlation decisions as well as classification of the fused inputs as specific events or activities. Variational inference provides the approximation necessary for reasoning over large scale data by decomposing the automated maximum likelihood estimation into perception and control, and allowing learning disentangled representations that are essential for generalizing across different domains. Finally, we will use the concepts from inverse reinforcement learning to update the parameters of common joint multi-modal representation. We will conclude this paper by studying the application of proposed model for two use-cases: construction of hypotheses and patterns of life from multi-media data where the data modalities contain text and imagery artifacts, and distributed decision-making in the geospatial environment where the input data can come from different overlapping and potentially conflicting observations from distinct observers, and contain information at different conceptual level, such as entity movement, skills, goals, and reactions.

Proceedings Article | 7 June 2018 Presentation + Paper

Optimizing collaborative computations for scalable distributed inference in large graphs

Georgiy Levchuk, John Colonna-Romano

Proceedings Volume 10646, 106460O (2018) https://doi.org/10.1117/12.2305872

KEYWORDS: Analytics, Data communications, Distributed computing, Optimization (mathematics), Computing systems, Telecommunications, Mining, Machine vision, Computer vision technology, Social network analysis

Read Abstract +

Proceedings Article | 21 May 2018 Presentation + Paper

Active learning and structure adaptation in teams of heterogeneous agents: designing organizations of the future

Georgiy Levchuk, Adam Fouse, Krishna Pattipati, Daniel Serfaty, Robert McCormack

Proceedings Volume 10653, 1065305 (2018) https://doi.org/10.1117/12.2305875

KEYWORDS: Performance modeling, Reconnaissance, Sensors, Robotics, Machine learning, Artificial intelligence, Defense technologies, Social networks, Computer programming

Read Abstract +

Many novel DoD missions, from disaster relief to cyber reconnaissance, require teams of humans and machines with diverse capabilities and intelligence. To succeed, DoD planners organize available personnel and technologies into mission-based teams and organizations. Enabled by next generation of sensors, new ways to access information, increasing capabilities of robotic platforms, and advances in machine learning and artificial intelligence for distributed inference and control applications, the new types of teams are emerging that include autonomous collaborating human and machine agents. Developing models to extract highest potential from human-machine teaming is the defense technology of the future. While many empirical studies have demonstrated the benefits of alternative organizations, such as adaptive networks command and control structures, traditional computational team design solutions have mostly focused on teams of homogeneous agents (such as swarms or social networks), and simple problems (such as cooperative task allocation, geospatial movement, and collaborative decision making). Because machines and humans often have distinct and complementary skills, team members could perform different roles and have changing relations over time. To improve team performance, new solutions are needed to dynamically adapt team structure to better fit the tasks that a team executes. In this paper, we present a continuation of our work on adaptive self-organizing teams. Our model is based on team active inference, the model that describes the approximate inference as an iterative minimization of the free variational energy encoding the task performance and team process complexity. Our model provides the methodology for adapting the structure of heterogeneous organization in distributed manner, where the agents on the team make local decisions to change their roles and relations which are synchronized through explicit collaborative messages. The roles of agents are defined through decomposition of the generalized task types into groups, and assignment of these groups to agents. We obtain decomposition groups using variational clustering on the factor graph, which defines the contribution of the tasks and their dependencies on the team’s objective function. This clustering constructs regions in the factor graph that trade-off independence, work balancing, and the overlap to help optimized organization obtain globally-optimal solutions in distributed manner under communication uncertainties.

Proceedings Article | 10 May 2018 Presentation + Paper

Algebra for distributed collaborative semi-supervised classification of cyber activities

Georgiy Levchuk, John Colonna-Romano, Mohammed Eslami

Proceedings Volume 10652, 1065210 (2018) https://doi.org/10.1117/12.2305869

KEYWORDS: Data communications, Analytics, Machine learning, Data modeling, Matrices, Data centers, Computer security, Network security, Telecommunications, Data processing

Read Abstract +

In the last several years, the volume and diversity of cyber attacks on the U.S. commercial and government networks have increased dramatically, including malware, web attacks (e.g., drive-by downloads), zero-day exploits, and men in the middle (e.g., session hijacking). While many tools are available to attackers, cyber criminals increasingly rely on straightforward intrusion approaches (e.g., spear-phishing), employ vast distributed resources (botnets), and hide attack vectors via stepping stone attacks. Detecting such activities and infrastructure represent the most difficult challenge to cyber-security professionals, because these threats are often locally invisible at the isolated subnetworks. Cyber threat detection tools employed in the field today fail to deal with data volume, speed, and diversity of the cyber-attacks. Intrusion Detection Systems (IDS) are ineffective against novel threats, while anomaly-based methods generate large number of false alarms and are difficult to interpret. Supervised algorithms require curated labeled datasets to train their models which do not exist for novel attacks. Yet, the biggest challenge of these systems is a requirement that all of the data be available at a single global repository. The cost of maintaining global repository and associated computation infrastructure becomes unsustainable as the volume of cyber data collection increases. As threat detection solutions are deployed predominantly to analyze local traffic collected within and on the border of a single organization, these tools are unable to detect attacks that are locally invisible, such as attacks cross-cutting organizational boundaries. In this paper, we describe a new computational framework which will enable distributed enterprises to (a) perform local inference computations; and (b) collaborate using global messages and hybrid strategies to detect a wide range of global threats that are not locally visible. First, we present a matrix-based algebra that generalizes a wide range of machine learning algorithms to maximize the breadth of attack phenomena to be detected. We then derive a semi-supervised attack detection model that uses a hybrid collaboration with adaptive local and global computations at distributed repositories to detect global events when it is not possible to move all relevant data into a centralized location. Finally, we propose a feedback model to create active human-in-loop system which integrates cyber analysts into malicious behavior detection and pattern learning process by generating requests for annotation and result examination using small number of representative instances of anomaly and threat detection outcomes

Proceedings Article | 19 May 2017 Presentation + Paper

Application of graph-based semi-supervised learning for development of cyber COP and network intrusion detection

Georgiy Levchuk, John Colonna-Romano, Mohammed Eslami

Proceedings Volume 10206, 102060D (2017) https://doi.org/10.1117/12.2263543

KEYWORDS: Computer intrusion detection, Systems modeling, Instrument modeling

Read Abstract +

Proceedings Article | 15 May 2017 Presentation + Paper

Application of free energy minimization to the design of adaptive multi-agent teams

Georgiy Levchuk, Krishna Pattipati, Adam Fouse, Daniel Serfaty

Proceedings Volume 10206, 102060E (2017) https://doi.org/10.1117/12.2263542

KEYWORDS: Reconnaissance, Fusion energy, Machine learning, Adaptive control, Thermodynamics, Information fusion, Defense and security, Sensors, Artificial intelligence, Unmanned aerial vehicles

Read Abstract +

Proceedings Article | 3 May 2017 Presentation + Paper

Using soft-hard fusion for misinformation detection and pattern of life analysis in OSINT

Georgiy Levchuk, Charlotte Shabarekh

Proceedings Volume 10207, 1020704 (2017) https://doi.org/10.1117/12.2263546

KEYWORDS: Data modeling, Associative arrays, Computer programming, Ruthenium, Algorithm development, Missiles, Data communications, Telecommunications, Web 2.0 technologies, Data fusion, Information fusion, Reliability, Knowledge discovery

Read Abstract +

Proceedings Article | 17 May 2016 Paper

Collaborative mining of graph patterns from multiple sources

Georgiy Levchuk, John Colonna-Romanoa

Proceedings Volume 9842, 98420O (2016) https://doi.org/10.1117/12.2225965

KEYWORDS: Mining, Data mining, Data modeling, Data fusion, Machine learning, Mining, Data mining, Evolutionary algorithms, Expectation maximization algorithms, Signal processing, Image classification, Algorithm development

Read Abstract +

Proceedings Article | 12 May 2016 Paper

Function and activity classification in network traffic data: existing methods, their weaknesses, and a path forward

Georgiy Levchuk

Proceedings Volume 9850, 985004 (2016) https://doi.org/10.1117/12.2225949

KEYWORDS: Network security, Classification systems, Image classification, Profiling, Analytical research, Data modeling, Space operations, Inspection, Performance modeling, Expectation maximization algorithms

Read Abstract +

Proceedings Article | 18 June 2015 Paper

Collaborative mining and transfer learning for relational data

Georgiy Levchuk, Mohammed Eslami

Proceedings Volume 9494, 94940O (2015) https://doi.org/10.1117/12.2177603

KEYWORDS: Data modeling, Algorithm development, Mining, Data communications, Image classification, Biological research, Data mining, Associative arrays, Expectation maximization algorithms, Performance modeling

Read Abstract +

Proceedings Article | 21 May 2015 Paper

Issues and challenges of information fusion in contested environments: panel results

Erik Blasch, Ivan Kadar, Chee Chong, Eric Jones, Jorge Tierno, Laurie Fenstermacher, John Gorman, Georgiy Levchuk

Proceedings Volume 9474, 94740Q (2015) https://doi.org/10.1117/12.2177523

KEYWORDS: Sensors, Information fusion, Data fusion, Intelligence systems, Data modeling, Web 2.0 technologies, Sensor networks, Defense and security, Analytics, Data communications

Read Abstract +

Proceedings Article | 20 June 2014 Paper

Detecting misinformation and knowledge conflicts in relational data

Georgiy Levchuk, Matthew Jackobsen, Brian Riordan

Proceedings Volume 9091, 90910P (2014) https://doi.org/10.1117/12.2050842

KEYWORDS: Data modeling, Sensors, Analytical research, Distance measurement, Detection and tracking algorithms, Associative arrays, Information fusion, Aluminum, Data processing, Improvised explosive devices

Read Abstract +

Proceedings Article | 20 June 2014 Paper

Visualization of graphical information fusion results

Erik Blasch, Georgiy Levchuk, Gennady Staskevich, Dustin Burke, Alex Aved

Proceedings Volume 9091, 90910L (2014) https://doi.org/10.1117/12.2052892

KEYWORDS: Data modeling, Visualization, Information fusion, Video, Data fusion, Sensors, Optical tracking, Visual analytics, Roads, Information visualization

Read Abstract +

Proceedings Article | 18 June 2014 Paper

Classification of group behaviors in social media via social behavior grammars

Georgiy Levchuk, Lise Getoor, Marc Smith

Proceedings Volume 9097, 909707 (2014) https://doi.org/10.1117/12.2050823

KEYWORDS: Web 2.0 technologies, Social networks, Analytical research, Virtual reality, Military intelligence, Cyber sensing, Mining, Data modeling, Electromagnetic coupling, Information security

Read Abstract +

Proceedings Article | 22 May 2014 Paper

Analysis of large-scale distributed knowledge sources via autonomous cooperative graph mining

Georgiy Levchuk, Andres Ortiz, Xifeng Yan

Proceedings Volume 9119, 91190K (2014) https://doi.org/10.1117/12.2050836

KEYWORDS: Data modeling, Data communications, Data processing, Data analysis, Mining, Analytical research, Sensors, Detection and tracking algorithms, Clouds, Data mining

Read Abstract +

Proceedings Article | 22 May 2014 Paper

Patterns of life in temporal data: indexing and hashing for fast and relevant data retrieval

Matthew Jacobsen, Georgiy Levchuk, Mark Weston, Jennifer Roberts

Proceedings Volume 9119, 91190J (2014) https://doi.org/10.1117/12.2053422

KEYWORDS: Data modeling, Discrete wavelet transforms, Binary data, Data processing, Fourier transforms, Expectation maximization algorithms, Data conversion, Algorithm development, Performance modeling, Databases

Read Abstract +

Proceedings Article | 28 May 2013 Paper

Fusing video and text data by integrating appearance and behavior similarity

Georgiy Levchuk, Charlotte Shabarekh

Proceedings Volume 8751, 875107 (2013) https://doi.org/10.1117/12.2014878

KEYWORDS: Video, Data modeling, Semantic video, Video surveillance, Visualization, Image resolution, Motion models, Sensors, Video processing, Feature extraction

Read Abstract +

Proceedings Article | 28 May 2013 Paper

Reusing information for high-level fusion: characterizing bias and uncertainty in human-generated intelligence

Dustin Burke, Alan Carlin, Paul Picciano, Georgiy Levchuk, Brian Riordan

Proceedings Volume 8758, 87580B (2013) https://doi.org/10.1117/12.2016310

KEYWORDS: Data modeling, Information fusion, Sensors, Reliability, Data fusion, Situational awareness sensors, Systems modeling, Intelligent sensors, Data processing, Associative arrays

Read Abstract +

Proceedings Article | 23 May 2013 Paper

Learning and detecting coordinated multi-entity activities from persistent surveillance

Georgiy Levchuk, Matt Jacobsen, Caitlin Furjanic, Aaron Bobick

Proceedings Volume 8745, 87451L (2013) https://doi.org/10.1117/12.2014875

KEYWORDS: Data modeling, Electro optical modeling, Motion models, Detection and tracking algorithms, Surveillance, Computer programming, Sensors, Video, Expectation maximization algorithms, Ions

Read Abstract +

Proceedings Article | 15 May 2012 Paper

Exploratory joint and separate tracking of geographically related time series

Balakumar Balasingam, Peter Willett, Georgiy Levchuk, Jared Freeman

Proceedings Volume 8393, 83930M (2012) https://doi.org/10.1117/12.919980

KEYWORDS: Data modeling, Statistical analysis, Visualization, Imaging systems, Defense and security, Databases, Radar, Radar imaging, Time metrology, Statistical modeling

Read Abstract +

Proceedings Article | 4 May 2012 Open Access

Multi-attributed network discovery: learning suspicious patterns in social network data (Withdrawal Notice)

Georgiy Levchuk, Jennifer Roberts, Jared Freeman

Proceedings Volume 8402, 84020A (2012) https://doi.org/10.1117/12.919398

KEYWORDS: Social networks, Biomimetics

Read Abstract +

DOWNLOAD PAPER

Proceedings Article | 20 May 2011 Paper

Wide-threat detection: recognition of adversarial missions and activity patterns in Empire Challenge 2009

Georgiy Levchuk, Charlotte Shabarekh, Caitlin Furjanic

Proceedings Volume 8059, 80590F (2011) https://doi.org/10.1117/12.886435

KEYWORDS: Data modeling, Detection and tracking algorithms, Improvised explosive devices, Sensors, Associative arrays, Phase modulation, Signals intelligence, Reconnaissance, Motion models, Performance modeling

Read Abstract +

Proceedings Article | 15 April 2010 Paper

Activity and function recognition for moving and static objects in urban environments from wide-area persistent surveillance inputs

Georgiy Levchuk, Aaron Bobick, Eric Jones

Proceedings Volume 7704, 77040P (2010) https://doi.org/10.1117/12.849492

KEYWORDS: Detection and tracking algorithms, Video surveillance, Data modeling, Video, Algorithm development, Buildings, Electro optical modeling, Surveillance, Lab on a chip, Classification systems

Read Abstract +

Proceedings Article | 30 April 2009 Paper

Adversarial behavior recognition from layered and persistent sensing systems

Georgiy Levchuk, Olga Mendoza-Schrock, Wayne Shebilske

Proceedings Volume 7347, 73470W (2009) https://doi.org/10.1117/12.818676

KEYWORDS: Data modeling, Buildings, Video surveillance, Video, Visual process modeling, Detection and tracking algorithms, Sensors, Analytical research, Visualization, Image classification

Read Abstract +

Proceedings Article | 16 April 2008 Paper

Recognition of coordinated adversarial behaviors from multi-source information

Georgiy Levchuk, Djuana Lea, Krishna Pattipati

Proceedings Volume 6943, 694305 (2008) https://doi.org/10.1117/12.777150

KEYWORDS: Data modeling, Signal to noise ratio, Detection and tracking algorithms, Weapons, Algorithm development, Sensors, Analytical research, Signals intelligence, Intelligent sensors, Reconnaissance

Read Abstract +

Showing 5 of 28 publications

Conference Committee Involvement (19)

Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIII

22 April 2024 | National Harbor, Maryland, United States

Signal Processing, Sensor/Information Fusion, and Target Recognition XXXII

1 May 2023 | Orlando, Florida, United States

Signal Processing, Sensor/Information Fusion, and Target Recognition XXXI

4 April 2022 | Orlando, Florida, United States

Disruptive Technologies in Information Sciences VI

4 April 2022 | Orlando, Florida, United States

Signal Processing, Sensor/Information Fusion, and Target Recognition XXX

12 April 2021 | Online Only, Florida, United States

Disruptive Technologies in Information Sciences V

12 April 2021 | Online Only, Florida, United States

Signal Processing, Sensor/Information Fusion, and Target Recognition XXIX

27 April 2020 | Online Only, California, United States

Disruptive Technologies in Information Sciences IV

27 April 2020 | Online Only, California, United States

Signal Processing, Sensor/Information Fusion, and Target Recognition XXVIII

15 April 2019 | Baltimore, MD, United States

Disruptive Technologies in Information Sciences III

15 April 2019 | Baltimore, MD, United States

Disruptive Technologies in Information Sciences

17 April 2018 | Orlando, FL, United States

Signal Processing, Sensor/Information Fusion, and Target Recognition XXVII

16 April 2018 | Orlando, FL, United States

Disruptive Technologies in Sensors and Sensor Systems

11 April 2017 | Anaheim, CA, United States

Signal Processing, Sensor/Information Fusion, and Target Recognition XXVI

10 April 2017 | Anaheim, CA, United States

Machine Intelligence and Bio-inspired Computation: Theory and Applications X

20 April 2016 | Baltimore, MD, United States

Signal Processing, Sensor/Information Fusion, and Target Recognition XXV

18 April 2016 | Baltimore, MD, United States

Machine Intelligence and Bio-inspired Computation: Theory and Applications IX

21 April 2015 | Baltimore, MD, United States

Signal Processing, Sensor/Information Fusion, and Target Recognition XXIV

20 April 2015 | Baltimore, MD, United States

Machine Intelligence and Bio-inspired Computation: Theory and Applications VIII

8 May 2014 | Baltimore, MD, United States

Showing 5 of 19 Conference Committees

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years