Yonghao Tan

• Address: HKUST, Clear Water Bay, Hong Kong, China
• Phone: +852 53946864
• E-mail: ytanaz@connect.ust.hk, yonghaot1017@gmail.com
• Personal Homepage: https://yonghao-tan.github.io/

Education & Academic Performance

• Ph.D. in Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST), Hong Kong, China (Sept. 2023 - Present)
  Supervised by Prof. Tim Kwang-Ting CHENG.
• B.E. in Microelectronics, Southern University of Science and Technology (SUSTech), Shenzhen, Guangdong, China (Sept. 2019 - Jun. 2023)
  Supervised by Prof. Fengwei An.
• Overall GPA: 3.77 / 4.0
• Weighted Average Score: 90.38
• Rank: 11 / 77

Research Interests

• Software/hardware co-design
• Model compression
• 3D processing

Research Experience

SLAM-Based Co-Design Embodied AI Chip

• Date: Jan. 2025 - Present
• Affiliation: AI Chip Center for Emerging Smart Systems, Hong Kong, China
• Mentor: Prof. Tim CHENG Kwang-Ting
• Integrate SLAM into a novel embodied AI system.
• Co-design exploration of heterogenous SLAM and AI computing core design.

Hybrid Bonding Based Co-Design AI Accelerator

• Date: Apr. 2024 - Present
• Affiliation: AI Chip Center for Emerging Smart Systems, Hong Kong, China
• Mentor: Prof. Tim CHENG Kwang-Ting
• Co-design optimization for LLMs.
• Implement the AI-core and ReRAM with 55nm die-on-wafer stacking via bumping process.

Transformer Based Co-Design AI Accelerator

• Date: Nov. 2021 - Sept. 2024
• Affiliation: AI Chip Center for Emerging Smart Systems, Hong Kong, China
• Mentor: Prof. Tim CHENG Kwang-Ting
• Hardware/software collaborative optimization of Transformer-based architecture.
• Implement an energy-efficient Transformer-based accelerator for semantic segmentation with 28nm ASIC process.

ASIC Design of SLAM Accelerator in 28nm CMOS Technology

• Date: Oct. 2021 - Jun. 2023
• Affiliation: Southern University of Science and Technology, Shenzhen, China
• Mentor: Prof. Fengwei An
• Propose a reconfigurable visual-inertial odometry accelerator and implement it on FPGA.
• Process data from image sensor and inertial measurement unit for real-time trajectory output at 160MHz and 110fps.
• Optimize hardware architecture and perform back-end design for ASIC development.

Reconfigurable SLAM Coprocessor

• Date: Jan. 2021 - Oct. 2021
• Affiliation: Southern University of Science and Technology, Shenzhen, China
• Mentor: Prof. Fengwei An
• Investigate and co-optimize SLAM algorithms and their operations.
• Propose a reconfigurable coprocessor with an instruction set that supports full functionality of operations in SLAM algorithms.

Publications

• 2025 Dong, P.*, Tan, Y.*, Liu, X., Luo, P., Liu, Y., Liang, L., … & Cheng, K. T.
  A 28nm 0.22uJ/Token Memory-Compute-Intensity-Aware CNN-Transformer Accelerator with Hybrid-Attention-Based Layer-Fusion and Cascaded Pruning for Semantic-Segmentation.
  In 2025 IEEE International Solid-State Circuits Conference (ISSCC), IEEE.

• 2024 Dong, P.*, Tan, Y.*, Zhang, D., Ni, T., Liu, X., Liu, Y., … & Cheng, K. T.
  Genetic Quantization-Aware Approximation for Non-Linear Operations in Transformers.
  In 2024 61st ACM/IEEE Design Automation Conference (DAC), IEEE.

• 2022 Tan, Y.*, Deng, H.*, Sun, M., Zhou, M., Chen, Y., Chen, L., … & An, F.
  A Reconfigurable Coprocessor for Simultaneous Localization and Mapping Algorithms in FPGA.
  IEEE Transactions on Circuits and Systems II: Express Briefs, 70(1), 286-290.

• 2022 Tan, Y.*, Sun, M.*, Deng, H., Wu, H., Zhou, M., Chen, Y., … & An, F.
  A Reconfigurable Visual-Inertial Odometry Accelerated Core with High Area and Energy Efficiency for Autonomous Mobile Robots.
  Sensors, 22(19), 7669.

* Both authors contributed equally to this research.

Honors and Awards

• Oct. 2022: First-class Outstanding Students Scholarship (Top 5% in SUSTech)
• May 2022: Successful Participant in Mathematical Contest in Modeling
• Dec. 2021: Shenzhen Longsys Electronics Company Award (Top 2% in School of Microelectronics)
• Dec. 2021: First Prize, 2021 National College Students FPGA Innovation Design Competition (Top 22 in 1341 teams)
• Oct. 2021: First Prize, 2021 International Competition of Autonomous Running Robots (Top 1 of 34 teams in final match)
• Oct. 2021: Second-class Outstanding Students Scholarship
• Mar. 2021: Third Prize, 2020 Asia and Pacific Mathematical Contest in Modeling
• Oct. 2020: Second-class Outstanding Students Scholarship

Fundings

• Sept. 2023 - Present: Postgraduate Studentship (PGS) Award in HKUST
• Apr. 2022: Undergraduate Innovation and Entrepreneurship Training Programs (Provincial Level)
• Jul. 2021: Guangdong College Students’ Scientific and Technological Innovation (Provincial Level)

Skills

• Programming language: C, C++, Java, Python, System Verilog, Verilog HDL, VHDL
• Professional software: AutoCAD, Cadence, Design Compiler, IC Compiler II, MATLAB, Multisim, Silvaco

Languages

• English (fluent), Mandarin (native), Cantonese (native)