Adversarial data poisoning attacks against the PC learning algorithm

Abstract

Data integrity is a key component of effective Bayesian network structure learning algorithms, namely PC algorithm, design and use. Given the role that integrity of data plays in these outcomes, this research demonstrates the importance of data integrity as a key component in machine learning tools in order to emphasize the need for carefully considering data integrity during tool development and utilization. To meet this purpose, we study how an adversary could generate a desired network with the PC algorithm. Given a Bayesian network B1 and a database DB1 generated by B1 and a second Bayesian network, B2, which is equal to B1, except for a minor change like a missing link, a reversed link, or an additional link, we explore and analyze what is the minimal number of changes such as additions, deletions, substitutions to DB1 that lead to a database DB2 that, when given as input to PC algorithm, results in B2.