REU Site on Integrated Machine Learning Systems:
Machine Learning and Law


Emily Meazell Amy McGovern

Andrew H. Fagg Dean F. Hougen

Examining existing legal framework for viability of integrating ML into real-world applications (Meazell, McGovern, Fagg, Hougen). The United States legal system conceptualizes, perceives, and responds to risk and uncertainty in particular ways that do not necessarily comport with the state of the art of ML applications. This disconnect reflects the differing values and goals of legal and scientific or engineering institutions; however, it can also function as a roadblock to effectively integrating ML into real-world applications. Thus, a better understanding of those differences and the reasons for them will both illuminate normative needs in the legal system, and suggest future research needs for science and engineering. This project will compare observed uncertainties and margins of error in an ML application of the students' choice, with legal doctrine, statutes, and regulations, to identify areas in which: (a) legal assumptions are incorrect or do not reflect state of the art; and (b) legal information needs are not being met such that ML is not a viable option. The students will develop recommendations for overcoming these areas with the goal of making ML fit comfortably within the legal system.

REU Home
Back to Projects

fagg [[at]]

Last modified: Wed Feb 20 00:04:13 2008