Research Areas

Digital technologies, powered by data-driven systems and operating on complex hardware, increasingly influence and govern our lives. We study the impacts of these consequential socio-technical systems on individuals, organizations, and societies. Our research aims to reveal the limitations of computational systems in addressing real-world problems. We seek to uncover the potential harms of such technologies and develop novel methods to mitigate them. Our vision is to create technologies that respect human values and avoid harming vulnerable and marginalized populations.

Cryptography is the foundation for secure communication. Our research is at the intersection of mathematics, computer science, and electrical engineering and is concerned with algorithms for tasks such as encryption or authentication, and more generally for securing digital information. One major direction of our research is the design, analysis, and secure implementation of cryptographic schemes that are secure against attacks by future large quantum computers, referred to as post-quantum cryptography. Another direction is using formal methods to investigate techniques and tools for high-assurance cryptography, i.e., schemes and implementations whose security is guaranteed by computer-verified proofs.

Privacy and data protection are paramount in the digital world, where the monitoring of personal information is ubiquitous. While laws and regulations are useful for mandating the protection of personal data, they can also cause unintended consequences that threaten the privacy of the user. Our vision is to enhance privacy under the threats of emerging technologies by engineering privacy-preserving systems, computationally operationalizing principles of data protection, and embedding end-users’ privacy needs in the development of systems and regulations.

We secure the world's digital infrastructure against cyberattacks. To achieve our aim, we employ a two-fold approach. On the one hand, we design and build secure computer systems, including software, hardware, and distributed systems. On the other hand, we examine the security of existing technologies, by developing automatic tools that have been deployed for widely-used systems. This work has led to the discovery and remediation of hundreds of thousands of security-critical weaknesses.

Global challenges today are complex and interconnected. To solve them, insights from one field are not enough. Our mission is to develop cross-disciplinary partnerships between data science and artificial intelligence (AI) research. We aim to develop robust and advanced models that oversee the diverse dimensions of societal safety. We tackle critical issues such as misinformation, bias, fraud, poverty, and disaster damage, ensuring a comprehensive safeguarding of our communities. We also build new methods to advance AI in other fields, particularly in neuroscience, protein engineering, and climate modeling.

This area is concerned with giving mathematically precise security definitions and verifying that systems are secure with respect to these definitions. To achieve this aim, formal methods tools such as automatic static analysis, model-checking, symbolic execution, and property-based testing are used. Additionally, we employ highly expressive but more interactive verification tools and proof assistants. These tools have solid foundations in logic and programming languages research and are applied to a diverse set of domains, including high-assurance systems, cryptography, differential privacy, secure compilation, smart contracts and blockchains, machine learning, etc.