Paulo Shakarian

Last updated

Paulo Shakarian
PhD
NationalityAmerican
Alma mater U.S. Military Academy, University of Maryland
Occupation(s) Associate professor, author
Known for Artificial intelligence, Cyber security
Honours Bronze Star, Army Commendation Medal

Paulo Shakarian is an associate professor at Arizona State University where he leads Lab V2 [1] which is focused on neurosymbolic artificial intelligence. His work on artificial intelligence and security has been featured in Forbes, the New Yorker, Slate, the Economist, Business Insider, TechCrunch, CNN and BBC. [2] [3] He has authored numerous books on artificial intelligence and the intersection of AI and security. He previously served as a military officer, had experience at DARPA, and co-founded a startup.

Contents

Scientific Work

Current Work

PyReason

In 2023, Shakarian's group released PyReason [4] which is a modern implementation of annotated logic [5] with extensions to support temporal and open-world reasoning. PyReason was used in various collaborations with industry partners. This included work with SSCI where PyReason was used as a "semantic proxy" to replace a simulation for reinforcement learning [6] where it provides a 1000x speedup over native simulation environments for agent policy training and provided transfer of PyReason-trained policies to simulation environments such as AFSIM and SC2. PyReason was also demonstrated as a method for robotic control in a joint ASU-SSCI demonstration. [7] In a separate line of work, under the IARPA HAYSTAC program [8] PyReason was used in a strategy to generate movement trajectories using ideas from abductive inference. [9] Here the authors leveraged properties of logic programming and A* search to generate movement trajectories that met certain criteria but resembled past agent activity.

Earlier Work

Social Network Diffusion

In the 2012 paper “Large social networks can be targeted for viral marketing with small seed sets”, [10] Shakarian introduced a fast, novel method for identifying sets of nodes that can maximize the spread of a contagion in a social network based on the standard “tipping model.” The work was presented in a 2012 ASONAM paper (later extended in a 2013 journal SNAM [11] and described in a 2015 book published by Springer-Nature [12] ). The concept was based around a graph decomposition designed to mimic the inverse of the diffusion process. The work was featured as part of MIT Technology Review’s “Best of 2013” and heralded as solving a “fundamental problem of viral marketing.” [13]

AI for Predicting Hacker Actions

In 2016, Shakarian’s team introduced a data mining framework in the paper “Darknet and deepnet mining for proactive cybersecurity threat intelligence” (Proc. IEE ISI 2016 [14] and later described in a book published by Cambridge University press in 2017 [15] ) which presented a framework for mining over 40 hacker websites – which not only demonstrated a scalable system for darkweb mining of hacker information, but also allowed for the ability to cross-examine cyber threat actors across multiple online forums – the study identified hundreds of hacker personas who participated in more than three different online marketplaces. The paper became one of the most cited papers of the history of the IEEE ISI conference and received media attention in Forbes [16] and MIT Technology Review. [17]

The following year, Shakarian and his team showed that data gathered from hacker communities on the dark web about specific software vulnerabilities often appeared before the use of zero-day exploits in a paper entitled “Proactive identification of exploits in the wild through vulnerability mentions online”. [18] [19] They found that that this information could also be used to create features for machine learning approaches can successfully predict the use of exploits – even when accounting for temporal intermixing of data. The approach was enhanced with follow-on studies were the features were augmented using social network topology data (Proc. ACM CSS 2017 [20] ) and the use of language models (Proc. AAAI 2018 [21] ).

Career

Shakarian was a major in the U.S Army serving from 2002 to 2014, undertaking two combat tours in Iraq and earning a Bronze Star and the Army Commendation Medal for valor. [3] [22] While in the army he was trained in Information assurance and completed a bachelor's degree in computer science at the U.S. Military Academy. [2] [22] In 2007 he served as a military fellow at Defense Advanced Research Projects Agency (DARPA). While in uniform, he went on to study a master's degree in computer science at the University of Maryland in 2009, and later a PhD in 2011 under the advisement of V.S. Subrahmanian. [23] His Ph.D. was focused on symbolic artificial intelligence, in particular logic programming, temporal logic, and abductive inference. [24]

After obtaining a PhD he taught at the U.S Military Academy, West Point, as an assistant professor from 2011 to 2014, his final military assignment. [25] In 2014 he took a position as an assistant professor at Arizona State University. [22] He earned his tenure at Arizona State and was promoted to Associate Professor in 2020. [26]

Since 2011 Shakarian has authored six books on subjects relating to his academic career - many of them focused on the intersection between AI, security, and data mining. [27]

In 2017, while maintaining his academic position he co-founded and led (as CEO) Cyber Reconnaissance, Inc., (CYR3CON), a business that specialized in combining artificial intelligence with information mined from malicious hacker communities to avoid cyber-attacks. [2] The company raised $8 million in venture capital [28] and was acquired in 2022. [29]

Notable works

Books

Related Research Articles

Artificial intelligence (AI), in its broadest sense, is intelligence exhibited by machines, particularly computer systems. It is a field of research in computer science that develops and studies methods and software that enable machines to perceive their environment and use learning and intelligence to take actions that maximize their chances of achieving defined goals. Such machines may be called AIs.

In artificial intelligence, symbolic artificial intelligence is the term for the collection of all methods in artificial intelligence research that are based on high-level symbolic (human-readable) representations of problems, logic and search. Symbolic AI used tools such as logic programming, production rules, semantic nets and frames, and it developed applications such as knowledge-based systems, symbolic mathematics, automated theorem provers, ontologies, the semantic web, and automated planning and scheduling systems. The Symbolic AI paradigm led to seminal ideas in search, symbolic programming languages, agents, multi-agent systems, the semantic web, and the strengths and limitations of formal knowledge and reasoning systems.

<span class="mw-page-title-main">Cybercrime</span> Type of crime based in computer networks

Cybercrime encompasses a wide range of criminal activities that are carried out using digital devices and/or networks. These crimes involve the use of technology to commit fraud, identity theft, data breaches, computer viruses, scams, and expanded upon in other malicious acts. Cybercriminals exploit vulnerabilities in computer systems and networks to gain unauthorized access, steal sensitive information, disrupt services, and cause financial or reputational harm to individuals, organizations, and governments.

The expression computational intelligence (CI) usually refers to the ability of a computer to learn a specific task from data or experimental observation. Even though it is commonly considered a synonym of soft computing, there is still no commonly accepted definition of computational intelligence.

Vasant G. Honavar is an Indian-American computer scientist, and artificial intelligence, machine learning, big data, data science, causal inference, knowledge representation, bioinformatics and health informatics researcher and professor.

<span class="mw-page-title-main">Richard Evans (AI researcher)</span>

Richard Evans is an artificial intelligence (AI) research scientist at DeepMind. His research focuses on integrating declarative interpretable logic-based systems with neural networks, and on formal models of Kant's Critique of Pure Reason.

General game playing (GGP) is the design of artificial intelligence programs to be able to play more than one game successfully. For many games like chess, computers are programmed to play these games using a specially designed algorithm, which cannot be transferred to another context. For instance, a chess-playing computer program cannot play checkers. General game playing is considered as a necessary milestone on the way to artificial general intelligence.

Drew McDermott was a professor of Computer Science at Yale University. He was known for his contributions in artificial intelligence and automated planning.

IEEE Intelligent Systems is a bimonthly peer-reviewed academic journal published by the IEEE Computer Society and sponsored by the Association for the Advancement of Artificial Intelligence (AAAI), British Computer Society (BCS), and European Association for Artificial Intelligence (EurAI).

Hsinchun Chen is the Regents' Professor and Thomas R. Brown Chair of Management and Technology at the University of Arizona and the Director and founder of the Artificial Intelligence Lab. He also served as lead program director of the Smart and Connected Health program at the National Science Foundation from 2014 to 2015. He received a B.S. degree from National Chiao Tung University in Taiwan, an MBA from SUNY Buffalo and an M.S. and Ph.D. in Information Systems from New York University.

<span class="mw-page-title-main">Eric Horvitz</span> American computer scientist, and Technical Fellow at Microsoft

Eric Joel Horvitz is an American computer scientist, and Technical Fellow at Microsoft, where he serves as the company's first Chief Scientific Officer. He was previously the director of Microsoft Research Labs, including research centers in Redmond, WA, Cambridge, MA, New York, NY, Montreal, Canada, Cambridge, UK, and Bangalore, India.

The cyber-arms industry are the markets and associated events surrounding the sale of software exploits, zero-days, cyberweaponry, surveillance technologies, and related tools for perpetrating cyberattacks. The term may extend to both grey and black markets online and offline.

Cyber threat intelligence (CTI) is a subfield of cybersecurity that focuses on the structured collection, analysis, and dissemination of data regarding potential or existing cyber threats. It provides organizations with the insights necessary to anticipate, prevent, and respond to cyberattacks by understanding the behavior of threat actors, their tactics, and the vulnerabilities they exploit. Cyber threat intelligence sources include open source intelligence, social media intelligence, human Intelligence, technical intelligence, device log files, forensically acquired data or intelligence from the internet traffic and data derived for the deep and dark web.

<span class="mw-page-title-main">Pedro Domingos</span> Professor of computer science

Pedro Domingos is a Professor Emeritus of computer science and engineering at the University of Washington. He is a researcher in machine learning known for Markov logic network enabling uncertain inference.

This glossary of artificial intelligence is a list of definitions of terms and concepts relevant to the study of artificial intelligence, its sub-disciplines, and related fields. Related glossaries include Glossary of computer science, Glossary of robotics, and Glossary of machine vision.

<span class="mw-page-title-main">Thomas G. Dietterich</span> American computer scientist and academic

Thomas G. Dietterich is emeritus professor of computer science at Oregon State University. He is one of the pioneers of the field of machine learning. He served as executive editor of Machine Learning (journal) (1992–98) and helped co-found the Journal of Machine Learning Research. In response to the media's attention on the dangers of artificial intelligence, Dietterich has been quoted for an academic perspective to a broad range of media outlets including National Public Radio, Business Insider, Microsoft Research, CNET, and The Wall Street Journal.

Karen L. Myers is the director of SRI International's Artificial Intelligence Center, where she is also principal scientist.

Neuro-symbolic AI is a type of artificial intelligence that integrates neural and symbolic AI architectures to address the weaknesses of each, providing a robust AI capable of reasoning, learning, and cognitive modeling. As argued by Leslie Valiant and others, the effective construction of rich computational cognitive models demands the combination of symbolic reasoning and efficient machine learning. Gary Marcus argued, "We cannot construct rich cognitive models in an adequate, automated way without the triumvirate of hybrid architecture, rich prior knowledge, and sophisticated techniques for reasoning." Further, "To build a robust, knowledge-driven approach to AI we must have the machinery of symbol manipulation in our toolkit. Too much useful knowledge is abstract to proceed without tools that represent and manipulate abstraction, and to date, the only known machinery that can manipulate such abstract knowledge reliably is the apparatus of symbol manipulation."

Yixin Chen is a computer scientist, academic, and author. He is a professor of computer science and engineering at Washington University in St. Louis.

References

  1. "Lab V2". labs.engineering.asu.edu.
  2. 1 2 3 CYR3CON.AI. "About". www.cyr3con.ai. Retrieved August 7, 2019.{{cite web}}: CS1 maint: numeric names: authors list (link)
  3. 1 2 "Paulo Shakarian". New America. Retrieved August 7, 2019.
  4. Aditya, D. (2023). "PyReason: Software for Open World Temporal Logic". AAAI-Make. arXiv: 2302.13482 .
  5. Kifer, Michael; Subrahmanian, V.S. (1992). "Theory of generalized annotated logic programming and its applications". Journal of Logic Programming. doi:10.1016/0743-1066(92)90007-P.
  6. Mukherji, K.; Parkar, D.; Pokala, L.; Aditya, D.; Shakarian, P.; Dorman, C. (2024). "Scalable Semantic Non-Markovian Simulation Proxy for Reinforcement Learning". IEEE Icsc. arXiv: 2310.06835 .
  7. "PyReason Sim-to-Real Demo". December 11, 2023 via YouTube.
  8. "HAYSTAC". www.iarpa.gov.
  9. Bavikadi, D.; et al. (2024). "Geospatial Trajectory Generation via Efficient Abduction: Deployment for Independent Testing". 40th Intl. Conference on Logic Programming (ICLP).
  10. Shakarian, P.; Paulo, D. (2012). "Large Social Networks Can be Targeted for Viral Marketing with Small Seed Sets". 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining. pp. 1–8. arXiv: 1205.4431 . doi:10.1109/ASONAM.2012.11. ISBN   978-1-4673-2497-7.
  11. Shakarian, Paulo; Eyre, Sean; Paulo, Damon (December 1, 2013). "A scalable heuristic for viral marketing under the tipping model". Social Network Analysis and Mining. 3 (4): 1225–1248. doi:10.1007/s13278-013-0135-7 via Springer Link.
  12. Diffusion in Social Networks. SpringerBriefs in Computer Science. 2015. doi:10.1007/978-3-319-23105-1. ISBN   978-3-319-23104-4 via link.springer.com.
  13. "Best of 2013: US Military Scientists Solve the Fundamental Problem of Viral Marketing". MIT Technology Review.
  14. Nunes, Eric; Diab, Ahmad; Gunn, Andrew; Marin, Ericsson; Mishra, Vineet; Paliath, Vivin; Robertson, John; Shakarian, Jana; Thart, Amanda; Shakarian, Paulo (2016). "Darknet and deepnet mining for proactive cybersecurity threat intelligence". 2016 IEEE Conference on Intelligence and Security Informatics (ISI). pp. 7–12. arXiv: 1607.08583 . doi:10.1109/ISI.2016.7745435. ISBN   978-1-5090-3865-7.
  15. Robertson, John; Diab, Ahmad; Marin, Ericsson; Nunes, Eric; Paliath, Vivin; Shakarian, Jana; Shakarian, Paulo (September 24, 2017). Darkweb Cyber Threat Intelligence Mining. Cambridge University Press. doi:10.1017/9781316888513. ISBN   978-1-107-18577-7.
  16. Murnane, Kevin. "Machine Learning Goes Dark And Deep To Find Zero-Day Exploits Before Day Zero". Forbes.
  17. "Machine-Learning Algorithm Combs the Darknet for Zero Day Exploits, and Finds Them". MIT Technology Review.
  18. Almukaynizi, Mohammed; Nunes, Eric; Dharaiya, Krishna; Senguttuvan, Manoj; Shakarian, Jana; Shakarian, Paulo (2017). "Proactive identification of exploits in the wild through vulnerability mentions online". 2017 International Conference on Cyber Conflict (CyCon U.S.). pp. 82–88. doi:10.1109/CYCONUS.2017.8167501. ISBN   978-1-5386-2379-4.
  19. "Google Scholar". scholar.google.com.
  20. Almukaynizi, Mohammed; Grimm, Alexander; Nunes, Eric; Shakarian, Jana; Shakarian, Paulo (October 19, 2017). "Predicting Cyber Threats through Hacker Social Networks in Darkweb and Deepweb Forums". Proceedings of the 2017 International Conference of the Computational Social Science Society of the Americas. Association for Computing Machinery. pp. 1–7. doi:10.1145/3145574.3145590. ISBN   978-1-4503-5269-7 via ACM Digital Library.
  21. Tavabi, Nazgol; Goyal, Palash; Almukaynizi, Mohammed; Shakarian, Paulo; Lerman, Kristina (April 27, 2018). "DarkEmbed: Exploit Prediction with Neural Language Models". Proceedings of the AAAI Conference on Artificial Intelligence. 32 (1). doi:10.1609/aaai.v32i1.11428 via ojs.aaai.org.
  22. 1 2 3 "Paulo Shakarian | iSearch". isearch.asu.edu. Retrieved August 7, 2019.
  23. "Home | VS Subrahmanian". vssubrah.github.io.
  24. Shakarian, Paulo (2011). "Spatio-Temporal Reasoning about Agent Behavior". UMD dissertation.
  25. "Coming Home: This West Point grad is using AI and Big Data for national security". Business Insider .
  26. "Paulo Shakarian Associate Professor at Arizona State University". December 21, 2020.
  27. 1 2 3 4 5 6 7 "Books by Paulo Shakarian". www.amazon.com. Retrieved August 7, 2019.
  28. "AI Cyber Attack Prediction Platform CYR3CON Secures $8.2 Million Financing" (Press release). July 14, 2020.
  29. "Albuquerque cybersecurity firm acquires Arizona machine learning startup".
  30. Neuro Symbolic Reasoning and Learning.