Hello! 👋
I am Chris Molloy, Ph.D.
I am currently persuring my Master's in Financial Mathematics at The London School of Economics. Recently, I completed my Ph.D. In Deep Learning at Queen’s University in Kingston, Ontario at the L1NNA Laboratory. These areas, Finance and Deep Learning, reflect my love of mathematics, which is also why I chose to focus on math and statistics while achieving my undergraduate degree in Computer Science at Queen's. Throughout my career I had the pleasure of working in many different environments, from hedge fund R&D, robotics programming, security research, and even house construction. I have a wide variety of skills that I use to solve all problems that I see in front of me. If you would like a more detailed overview of my past work and education experience you may refer to my resume linked below.
View CVProjects
📷 Viz
In this work, we propose the first practical and efficient solution for zero-day malware variant matching with reconstruction, Viz. By combining multi-modality learning and a Siamese-based structure, our model can navigate across different modalities and match zero-day variants. To address the missing or noisy modality issue, we propose a Conditional Variable Autoencoder with a Generative Adversarial Network for heightened resolution. This work was published in the 2022 IEEE International Conference on Data Mining.The publication can be found here.
☯️ H4RM0NY
H4RM0NY is a two player game for malware generation and detection. H4RM0NY automates the process of obfuscating malware through a two party game using an OpenAI Gym. As its name suggests, H4RM0NY is not all evil, but we have seen that playing this game also strengthens malware detection networks against adversarial samples! This work was published in The 2022 IEEE International Conference on Cyber Security and Resilience. The publication can be found here. This paper was awarded the best research paper award.
🔬 JARV1S
JARV1S is a phenotype based malware decomposition and comparison platform. Phenotypes are decomposed from assembly functions and used to search for matching malware variants. Using phenotypes allows us to skip over time consuming sandboxing, as well as store small and searchable information about malware samples. JARV1S has been published in the 14th International Conference on Cyber Conflict. A demo video of JARV1S can be found here.
⛩️ Goshuin
In May 2023 I went on a trip to Japan with a group of friends. While there, my girlfriend and I fell in love with collecting temple stamps (御朱印). Temple stamps are written in a stamp book (御朱印帳) by monks or other artistis who work at that specific shrine. We traveled to Tokyo, Hakone, and Kyoto collecting stamps while site seeing, shopping, and having great food! Please see the Goshuin page for a documentation of the shrines we visited.
📚 Encyclopedia of Machine Learning
For the Encyclopedia of Machine Learning and Data Science I was given the opportunity to write a chapter about adversarial learning on malware. This was my first project in grad school and it was a great way to learn about the topic that I am currently researching. The chapter can be found here.
💉 Snitch
Snitch is the name of a network I designed for finding drug dealers on public forums. The network was designed with TensorFlow and there are promising results shown on the Wall Street dark net forums. This project is all open source and can be found here.
🪴 QVFT
Queen's Vertical Farming Team (QVFT) is a club at Queen's that I was a member of for 2 years. While I was a member of QVFT I was tasked with designing the implementing a database that would store plant information. I also aided in web design.
Awards & Certificates
NSERC Postgraduate Scholarships - Doctoral program, awarded by the Natural Sciences and Engineering Research Council of Canada.
Lab2Market Cybersecurity Mitacs Accelerate Certificate, awarded by Mitacs.
Best Research Paper Award, awarded by the 2022 IEEE Conference on Cybersecurity and Resiliance Chairs.
Udemy: Quantitative Finance & Algorithmic Trading in Python, certifies the completion of the Quantitative Finance & Alg. Trading in Python course
Udemy: Introduction to Stochastic Calculus Certificate, certifies the completion of the Introduction to Stochastic Calculus course.
Publications
C. Molloy, J. Banks, H. H. Steven Ding, P. Charland, A. Walenstein and L. Li, "Adversarial Variational Modality Reconstruction and Regularization for Zero-Day Malware Variants Similarity Detection," 2022 IEEE International Conference on Data Mining (ICDM), Orlando, FL, USA, 2022, pp. 1131-1136, doi: 10.1109/ICDM54844.2022.00143. [Preprint]
C. Molloy, P. Charland, S. H. H. Ding and B. C. M. Fung, "JARV1S: Phenotype Clone Search for Rapid Zero-Day Malware Triage and Functional Decomposition for Cyber Threat Intelligence," 2022 14th International Conference on Cyber Conflict: Keep Moving! (CyCon), Tallinn, Estonia, 2022, pp. 385-403, doi: 10.23919/CyCon55549.2022.9811078. [Preprint]
C. Molloy, S. H. H. Ding, B. C. M. Fung and P. Charland, "H4rm0ny: A Competitive Zero-Sum Two-Player Markov Game for Multi-Agent Learning on Evasive Malware Generation and Detection," 2022 IEEE International Conference on Cyber Security and Resilience (CSR), Rhodes, Greece, 2022, pp. 22-29, doi: 10.1109/CSR54599.2022.9850345. [Preprint]
Molloy, C., Mansour, Z., Ding, S.H.H. (2022). Adversarial Learning on Malware. In: Phung, D., Webb, G.I., Sammut, C. (eds) Encyclopedia of Machine Learning and Data Science. Springer, New York, NY. https://doi.org/10.1007/978-1-4899-7502-7_982-1 [Preprint]
Mansour, Z., Molloy, C., Ding, S.H.H. (2022). Machine Learning for Static Malware Analysis. In: Phung, D., Webb, G.I., Sammut, C. (eds) Encyclopedia of Machine Learning and Data Science. Springer, New York, NY. https://doi.org/10.1007/978-1-4899-7502-7_981-1
Li, L., Ding, S., Charland, P., Yu, H. and Molloy, C.J., GenTAL: Generative Denoising Skip-gram Transformer for Unsupervised Binary Code Similarity Detection.