The objective of this project is to develop strategy-aware learning algorithms for PAC learning and generalization from compromised data along with development of privacy preserving control policies for multiagent systems and obfuscator algorithm for privacy preservation