Song Han

Song Han is an Associate Professor in the Department of Electrical Engineering and Computer Science (EECS) at the Massachusetts Institute of Technology (MIT). His research focuses on efficient deep learning computing, bridging algorithm and hardware design to enable scalable AI systems.

• PhD in Electrical Engineering from Stanford University

Research Interests

• Efficient Deep Learning
• Neural Network Compression
• Hardware-Aware Transformers
• Sparse Attention Mechanisms
• Quantization Techniques
• Edge and IoT Computing

Recent Publication Trends highlight advances in LLM optimization, diffusion model quantization, and quantum-classical co-design. His work emphasizes reducing computational costs while maintaining model fidelity.

Scientific Awards

• Best Paper, ICLR and FPGA Symposium
• NSF CAREER Award
• MIT Technology Review 35 Innovators Under 35

Collaborations include the MIT-IBM Watson AI Lab, focusing on AI hardware and system co-design. Many of his techniques are integrated into commercial AI chips.