Border Patrol, an organization within Customs and Border Protection in the Department of Homeland Security, is responsible for securing the U.S. land border between official Ports of Entry. This responsibility includes the interdiction and apprehension of migrants attempting to enter the country illegally. In recent years, several Agent Based Models have been built to examine operational improvements, including sensor placement, patrol patterns, and estimation of total migrant flows. To date, none of these models have implemented intelligent agent interactions to allow agents to adapt future behaviors to previous experiences and success or failure in meeting key objectives. This paper describes an Agent Based Model that was built to demonstrate and test the effectiveness of several different learning strategies for Border Patrol and Migrant agents, and reports the results of learning strategy comparisons for both agent types.