Scalable machine learning technology, evolutionary algorithms, and data science frameworks for large-scale knowledge mining, prediction, analytics, and optimization with projects in cyber security, MOOC technology, software analysis, and clinical medicine



Research Themes

Artificial Adversarial Intelligence MOOC Data Analytics Clinical Medicine





Applications: Adversarial dynamics modeling, adversarial examples, adversarial hardening, malware detection and classification, automatic program analysis for vulnerabilities, deception, segmentation and peer to peer network defenses against DDOS, lateral stage and scanning attacks; artificial intelligence; MOOCs on computational thinking and programming, MOOC clickstream and activity data analysis with deep learning; Large scale physiological time series, EEG and other vital signals, leveraging deep learning and similarity-based retrieval.



ALFA's research would not be possible without the past and ongoing support of our industry sponsors
The views, opinions and positions expressed by ALFA Group and on this site are theirs alone, and do not necessarily reflect the views, opinions or positions of their sponsors
ALFA Group is formerly the Evolutionary Design and Optimization Group

Recent News


Dr. Jamal Toutouh is awarded the Fellow of the Month of the Marie Sklodowska Curie Alumni Association -North America Chapter


Flipping Sign Bits is All You Need to Craft Black-Box Adversarial Examples
Abdullah Al-Dujaili
ICML Security Vulnerabilities of Machine Learning Systems Workshop, Long Beach CA USA

We welcome Dr. Şebnem Özdemir, Assistant Professor from Istinye University, Istanbul Turkey


An Artificial Coevolutionary Framework for Adversarial AI
Erik Hemberg
MIT - Lincoln Lab, Cambridge MA USA

Artificial Adversarial Intelligence for Cybersecurity
Una-May O'Reilly
IBM, Tokyo Japan

We welcome Dhruv Sumathi and Alexander Shushkov, Interns from Weston High School


Spatial Evolutionary Generative Adversarial Network
Erik Hemberg
SystemsThatLearn@CSAIL, Research Update Report, Cambridge MA USA

We welcome a new Postdoctoral Fellow to the group Dr. Behnaz Bostanipour, from University of Lausanne, Switzerland, sponsored by the Postdoctoral Mobility Fellowship of the Swiss National Science Foundation.



Flipping Sign Bits is All You Need to Craft Black-Box Adversarial Examples
Abdullah Al-Dujaili
IBM Research, Cambridge MA USA

Technology Trends in Cybersecurity, Artificial Intelligence and Machine Learning
Una-May O'Reilly
Canadian Embassy of the USA, Washington DC USA

Adversarial Dynamics; Anticipicating The Next Move
Una-May O'Reilly
AI/ML Cyber Defense Ops, Workshop on Implications of AI and Cybersecurity, R&D and Emerging Areas Panel
NASEM AI and CyberSecurity Workshop, Washington DC USA

We welcome Sagar Biswas, visiting student from Manipal University, Jaipur India in collaboration with Hong Kong University, Hong Kong HK



Artificial Adversarial Intelligence
Una-May O’Reilly
AAAI-19, 33rd AAAI Conference on Artificial Intelligence, Honolulu HI USA




Computer Systems Arms Races
Una-May O’Reilly
Dalhousie University, Halifax Canada
Keynote Speaker: The ACM Canadian Celebration of Women in Computing, Halifax Canada



*Outstanding Poster Award: Machine Learning and Malware Arms Race
Abdullah Al-Dujaili, Alex Huang, Erik Hemberg, Una-May O'Reilly
2018 Horizons Colloquium, awarded on behalf of IBM Research
MIT-IBM Watson AI Lab Medium Article: Inaugural AI Research Week Highlights Collaboration and Future of AI
We welcome Carl Im, CEO, Solomon Strategic Consulting Limited, Seoul South Korea visiting to oberserve ALFA research


Data-driven Bug Hunting - Tales From Hunting Down Bugs in Ethereum's Solidity
Shashank Srikant
Systems That Learn @ CSAIL Annual  Meeting, Cambridge MA USA
Towards Robust and Scalable Generative Adversarial Networks
Una-May O’Reilly
Systems That Learn @ CSAIL Annual  Meeting, Cambridge MA USA
Computer Systems Arms Races
Una-May O’Reilly
Emtech MENA, Dubai UAE
Abdul Latif Jamell International, Dubai UAE
Parsons, Dubai UAE

We welcome a new group member, Jamal Toutouh, Post-doctoral Fellow from Universidad de Malaga, Spain



The Rise of Deep Learning
Una-May O’Reilly
Accenture, Webinar


Old News...