Associate Professor Yingjie Lao researches the intersection of artificial intelligence, security, and hardware. His lab develops Python code for Artificial Intelligence systems and hardware description language (HDL). He also explores VLSI architectures for machine learning and emerging cryptographic systems, cybersecurity, and robotics.
Lao's research projects include:
- Trusted and Responsible AI
- Hardware Security
- Acceleration for Security and Cryptographic Algorithms
Trusted and Responsible AI
Through the development of powerful algorithms and design tools, AI has approached or even surpassed human-level performance in many applications. However, as AI systems (e.g., large language models) become more integrated into critical aspects of society, the need for trusted and responsible AI becomes paramount. Ensuring these systems are transparent, ethical, and accountable is essential to prevent biases, protect privacy, and maintain public trust. This research seeks to enhance AI fairness, accountability, and transparency, as well as robust mechanisms for privacy and security, to ensure that AI technologies benefit all of society equitably and safely.
Hardware Security
With the rapid development and globalization of the semiconductor industry, hardware security has emerged as a critical concern due to the risks posed by an untrusted supply chain, where malicious components can be introduced during manufacturing and deployment. Additionally, the accessibility and unsupervised nature of hardware devices expose them to new attacking and tampering methods, continuously challenging current protection strategies. Therefore, securing next-generation systems from the hardware or physical perspective is of paramount importance. Research topics in hardware security include designing secure hardware architectures, developing techniques to detect and prevent hardware-level threats, and implementing robust cryptographic mechanisms to safeguard data.
Acceleration for Security and Cryptographic Algorithms
The increasing demand for data privacy and security has driven the need for efficient acceleration in emerging security and cryptographic algorithms, such as Fully Homomorphic Encryption (FHE) and Post-Quantum Cryptography (PQC). These cryptographic techniques, while robust, are computationally intensive, making acceleration essential for their practical deployment in real-world applications. Research topics in this area include algorithm-hardware co-optimization of these security and cryptographic algorithms, the design of specialized hardware accelerators, and the development of scalable architectures to reduce computational complexities.