Chang Eun (Paul), Song

CV | Contact | Linkedin

My research interests include HW-SW Co-Design of AI Processors, ML Accelerators design for large language models, In-memory computing, and Neuromorphic computing. I work on the end-to-end flow for AI processor design, from ML optimization to Chip fabrication and measurement.

I'm a third-year Ph.D. student in Computer Science and Engineering at the University of California, San Diego (UCSD), co-advised by Prof. Mingu Kang and Prof. Tajana Rosing.

I received a Bachelor's in Electrical Engineering from Korea University in February 2022.

<Skillset>

[Latest Updates!] (Click)

<Education>

2022.09 - Present

Ph.D. Student in Computer Science and Engineering  (Co-Advisor: Mingu Kang, Tajana Rosing)

2016.03 - 2022.02

Bachelor of Engineering in Electrical Engineering  (Two years absence for military service (2018-2019))

<Publications>

Conference Paper (* Equal Contribution)


[C.4] C. E Song, P. Bhatnagar, Z. Xia, T. Rosing, and M. Kang, "Hybrid Analog Processing in Memory for Transformer Accelerator" (Under Review)


[C.3] H. Yang*, C. E. Song*, W. Xu, B. Khaleghi, U. Mallappa, M. Shah, K. Fan, M. Kang, and T. Rosing, “End-to-end Few-shot Learning Accelerator with Integrated Feature Extraction and Hyperdimensional Computing Classifier”, IEEE European Solid-State Electronics Research Conference (ESSERC), 2024 [Link]


[C.2] C. E. Song*, A. Moradifirouzabadi*, T. Rosing, and M. Kang, “Efficient Transformer Acceleration via Reconfiguration for Encoder and Decoder

Models and Sparsity-Aware Algorithm Mapping”, ACM/IEEE International Symposium on Low Power Electronics and Design (ISLPED), 2024 [Link]


[C.1] C. E. Song, Y. Li, A. Ramnani, P. Agrawal, P. Agrawal, S. Jang, S. Lee, T. Rosing, and M. Kang, “52.5 TOPS/W 1.7GHz Reconfigurable XGBoost Inference Accelerator based on Modular-Unit-Tree with Dynamic Data and Compute Gating”, IEEE Custom Integrated Circuits Conference (CICC), 2024 [Link]

<Work/Research Experience>

2023.01 - Present | Graduate Student Researcher (Advisor: Prof. Tajana Rosing, http://varys.ucsd.edu/)

- AI Processor design for hyperdimensional computing, few-shot learning, and ReRAM in-memory computing

- Chip fabricated in TSMC N40 (one digital, and one mixed signal), including front-end (RTL design) and back-end (floorplanning, place & route, and gds2)

2022.09 - Present | Graduate Student Researcher (Advisor: Prof. Mingu Kang, https://ucsdvvip.com/)

- AI Processor design for large language model, and decision tree (XGBoost)

- Digital chip fabricated in TSMC N65, including ML optimization, front-end (RTL design), and back-end (floorplanning, place & route, and gds2)

2024.06 - 2024.09 | AI Hardware Research Intern (Manager: Kerem Akarvardar)

- Circuit and architecture co-design for softmax optimization in Large Language Model

2021.09 - 2021.12 | Undergraduate Student Researcher (Advisor: Prof. Park, Jong-Sun)

- “Enhanced Stability of 8T and 10T SRAM over 6T SRAM with monte carlo simulation”; Graduation Thesis II;

2021.06 - 2021.08 | Undergraduate Student Researcher (Advisor: Prof. Kim, Tony Tae-Hyoung)

- Research on Computing-In-Memory: Analyzed Ambit In-Memory Accelerator & Advanced Computer Architecture

2021.01 - 2021.06 | Undergraduate Student Researcher (Advisor: Prof. Lee, Hyung-Min)

- “Design of Low-Dropout (LDO) Linear Regulator ICs With Advanced Transient Response”; Graduation Thesis I;

<Patent>

[P.1] Changeun Song, “Method and System for Providing Virtual Reality Space”, 10-2020-0019618, Republic of Korea, 2020

<Honors & Awards>

<Chip Gallery>

XGBoost Accelerator 

in TSMC N65

(CICC'24)

FSL-HDnn

Accelerator

in TSMC N40

(ESSERC'24)

HD + ReRAM

Accelerator

in TSMC N40

(To be published soon)