Machine Learning: Lifelong Streaming Anomaly Detection
American University, Washington DC
Supervisors: Nathalie Japkowicz (CS), Roberto Corizzo (CS), Michael Baron (Statistics)
Description of the Work:
This work will take place in the context of DARPA’s L2M program (www.darpa.mil/program/lifelong-learning-machines). The successful candidate will expand American University team’s Lifelong Learning model and integrate it with SRI International’s model on both a Reinforcement Learning Task and unsupervised and one-class anomaly detection problems (in the context of cybersecurity, medical imaging, social media security, and many other applications). This work will be carried on in collaboration with Computer Science professors Nathalie Japkowicz and Roberto Corizzo and Statistics professor Michael Baron. The work will combine neural network (deep learning) and other machine learning techniques with statistical change-point detection methods. The resulting system will automatically recognize anomalies from stream data in real time, identify them as part of a concept drift, novelty, or accident/attack, and adapt to them according to their identification.
The ideal candidate will hold a Ph.D. in Computer Science in the area of Machine Learning and Neural Networks. The candidate must have an interest and some experience in a variety of machine learning techniques such as reinforcement learning, unsupervised or one-class learning, outlier detection, time-series analysis, and data stream learning techniques. The candidate must have experience or interest in integrating statistical methods in the system and be interested in collaborating with a team of statisticians. The candidate will have a record of publications in well-recognized conference and/or journal venues.
The principal duties to be carried out include:
• Integration of the current American University (AU) team’s Lifelong Learning approach with SRI International’s approach to the Starcraft application.
• Helping in the design and implementation of the new components of AU team’s approach.
• Adapting the approach to the various domains of application considered by the team.
• Thoroughly evaluating the performance of the system on these domains.
• Co-supervising Senior or Master’s students’ implementations (with testing and refinement) of sub-components or specific applications of the project.
• Preparing conference and journal manuscripts in collaboration with Drs. Japkowicz, Corizzo, and Baron and the students involved in the projects.
• Assisting in the preparation of additional grant proposals to continue the project or develop new ideas emanating from the project.
The successful candidate will be affiliated with the Department of Computer Science at American University in Washington, DC. The candidate will be working with Professors Nathalie Japkowicz, Roberto Corizzo, and other students affiliated with the lab. The candidate will also be working with Professor Michael Baron in the Statistics department.
The appointment is a 12-month term position (renewable for 12 months pending successful completion of the work and funding) and will commence on January 4, 2021, subject to the final budget approval.
Submitting an Application
Salary is competitive. Review of applications will begin immediately and will continue until the position is filled, subject to ongoing budgetary approval. Please submit applications via: http://apply.interfolio.com/79810. Include a letter of application, curriculum vita, three letters of recommendation, and copies of recent published papers or working papers. Please contact Nathalie Japkowicz, at firstname.lastname@example.org if you have any questions.
About the University
American University is a private institution within easy reach of the many centers of government, business, research, and the arts located within the nation’s capital. For more information about American University, visit www.american.edu.
American University is an equal opportunity, affirmative action institution that operates in compliance with applicable laws and regulations. The university does not discriminate on the basis of race, color, national origin, religion, sex (including pregnancy), age, sexual orientation, disability, marital status, personal appearance, gender identity and expression, family responsibilities, political affiliation, source of income, veteran status, an individual’s genetic information or any other bases under federal or local laws (collectively “Protected Bases”) in its programs and activities. American University is a tobacco and smoke free campus.