Upcoming Events

Center for Domain-Specific Computing (CDSC) Annual Review

February 24, 2025

Mong Auditorium, Engineering VI

February 25, 2025

Shannon Room, 54-134 Engineering IV

Day 1 (February 24, 2025)

Mong Auditorium, Engineering VI (No Food or Drink)

8:00am-8:30am Breakfast (First Floor Breezeway and Patio, Tannas Alumni Suite-176 Engineering VI)

8:30am-8:40am Greeting from the leadership from the School of Engineering, Alissa Park, Ronald and Valerie Sugar Dean, UCLA Samueli School of Engineering

8:40am-8:50am Greeting from the leadership from the School of Engineering, Todd Millstein, Department Chair, UCLA CS

8:50am-9:05am Welcome and CDSC Overview; Jason Cong, UCLA CS/ECE

9:05am–9:45am Keynote Speech 1: Gen AI Inference at the Edge — Challenges and Opportunities, Durga Malladi, Senior Vice President and General Manager, Technology Planning & Edge Solutions, Qualcomm Technologies, Inc.

9:45am-10:00am Break (First Floor Breezeway and Patio, Tannas Alumni Suite-176 Engineering VI)

Session 1: Energy-Efficient ML Models and Accelerations at the Edge (Chair: Jason Cong, UCLA CS/ECE)

10:00am-10:20am HMT: Hierarchical Memory Transformer for Long Context Language Processing (Zifan He, UCLA CS)

10:20am-10:40am Dynamic-Width Speculative Beam Decoding for Fast LLM Inference (Zongyue Qin, UCLA CS)

10:40am-11:00am Accelerating Large Language Model Pretraining via LFR Pedagogy: Learn, Focus, and Review (Neha Prakriya, UCLA CS)

11:00am-11:20am Improved Optimizers for LLM Training (Cho-Jui Hsieh, UCLA CS)

11:20am-11:40am HMT on an FPGA (Jiahao Zhang, UCLA CS)

11:40am-12:00pm Future of AI-driven Micro-mobility in Urban Spaces from Food Delivery Bots to Humanoid Robots (Bolei Zhou, UCLA CS)

12:00pm-1:20pm Lunch (First Floor Breezeway and Patio, Tannas Alumni Suite-176 Engineering VI)

Session 2: ML for Science, Chip Design, and Medicine (Chair: Yizhou Sun, UCLA CS)

1:20pm-1:40pm Foundation Models for Planetary-Scale Science (Aditya Grover, UCLA CS)

1:40pm-2:00pm Interacting Dynamical System Modeling for Science: Expressivity, Generalizability, and Uncertainty(Xiao Luo, UCLA CS)

2:00pm-2:20pm Hardware-Informed Machine Learning for High-Level Synthesis, Weikai Li, UCLA CS)

2:20pm-2:40pm Towards Generalizable Modeling of HLS Design with High Quality Synthetic Data (Brady Ding, UCLA CS)

2:40pm-3:00pm Vision Language Models and Applications in Radiology, (William Hsu, UCLA Radiological Sciences)

3:00pm-3:20pm Opportunities and Challenges for Foundation Models in Healthcare (Alex Bui, UCLA Radiological Sciences)

3:20pm-3:40pm Break (First Floor Breezeway and Patio, Tannas Alumni Suite-176 Engineering VI)

Session 3: Compilation for Customized Computing (Chair: Zhiru Zhang, Cornell ECE)

3:40pm-4:00pm Differentiable Optimization for Accelerator Compilation (Zhiru Zhang, Cornell ECE)

4:00pm-4:20pm StreamHLS: Towards Automatic Dataflow Acceleration (Suhail Basalama, UCLA CS)

4:20pm-4:40pm Automatic Code Generation for High-Level Synthesis through Nonlinear Programming-Based Space Exploration (Stéphane Pouget, UCLA CS)

4:40pm-5:00pm NoH: NoC Compilation in High Level Synthesis (Jake Ke, UCLA CS)

5:00pm-5:20pm Correctness Verification of Program Transformations (Louis-Noël Pouchet, Colorado State)

5:20pm-5:40pm Poster Introductions

6:00pm-8:00pm Reception + Poster Session (First Floor Breezeway and Patio, Tannas Alumni Suite-176 Engineering VI)

Day 2 (February 25, 2024)

Shannon Room, 54-134 Engineering IV (No Food or Drink)

8:00am-8:30am Breakfast (Tesla Room, 53-125 Engineering IV)

Session 4: Acceleration at the edge (Chair: Tony Nowatzki, UCLA CS)

8:30am-9:00am Energy is Everything: Towards General Purpose Computing at 1 TOPS/Watt, Nathan Beckman, CMU (Invited talk)

9:00am-9:20pm Addressing Architectural Obstacles for Overlay with Stream Network Abstraction (Chengyue Wang, UCLA CS)

9:20am-9:40am Understanding Fundamental Challenges of Dataflow Acceleration on General-Purpose Programs (Chris Liu, UCLA CS)

9:40am-10:00am SAT-Accel: HW Acceleration of Modern SAT Solvers (Michael Lo, UCLA CS)

10:00am-10:20am Multi-FPGA Acceleration for Large Scale Conjugate Gradient Solvers (Linghao Song, UCLA CS and Yale Univ.)

10:20am-10:40am Coffee Break (Tesla Room, 53-125 Engineering IV)

Session 5: Quantum Computing and Compilation (Chair: Jason Cong, UCLA CS/ECE)

10:40am-11:10am Quantum Computing and Ansys, Robert Lucas, Ansys Fellow, (Invited talk)

11:10am-11:30am Optimality Study of Quantum Layout Synthesis and Scalable Solutions (Shuohao Ping and Wan-Hsuan Lin, UCLA CS)

11:30am-11:50pm ZAC: Reuse-Aware Compilation for Zoned Quantum Architectures (Wan-Hsuan Lin, UCLA CS)

11:50pm-12:10pm QSP Circuit Optimization Exploiting Algebraic and Boolean Methods (Hanyu Wang, UCLA CS)

12:10pm-1:30pm Lunch (Tesla Room, 53-125 Engineering IV)

Session 6: Feedbacks from Industry Partners

1:30pm-2:30pm Feedbacks from Industry Partners

BIOs of the keynote speaker and invited speakers

Dr. Durga Malladi is Senior Vice President and General Manager,Technology Planning & Edge Solutions, at Qualcomm Technologies, Inc. He joined as a Senior Engineer in 1998. Since 2023, he has been responsible for technology product management and roadmap planning across all businesses in Qualcomm Technologies. This spans artificial intelligence (hardware, software, tools), connectivity (5G, Wi-Fi and Bluetooth, satellite communications, positioning), processors (CPU, GPU, NPU), multimedia (computer vision, audio, video, sensors), central software, developer ecosystem, and data management and analytics platforms. In addition, he is responsible for Qualcomm’s cellular infrastructure business. In prior roles, Durga led wireless research in 5G and 4G LTE/LTE-Advanced, and drove the 5G Modem-RF technology roadmap and mobile broadband business in Qualcomm. Durga is a senior member of IEEE and holds 581 U.S. patents. He is a recipient of Qualcomm’s IP Excellence Award, Qualcomm Distinguished Contributor Award for Project Leadership, Qualcomm Upendra Patel Achievement Awards for Outstanding Contributions, and Distinguished Alumnus Award from Indian Institute of Technology, Madras. He is a member of the AI Governance Alliance Steering Committee at the World Economic Forum. Durga holds a B.Tech (’93) from Indian Institute of Technology, Madras, M.S (’95) and Ph.D. (’98) from UCLA, and an AI Graduate Certificate from Stanford (’23). His Ph.D. dissertation is on adaptive estimation and filtering techniques, and his research interests include artificial intelligence, signal processing, communication theory, and quantum computing.

Dr. Robert F. Lucas is an Ansys Fellow where he is responsible for the default multifrontal linear solver used in LS-DYNA and MAPDL. Previously, he was the Operational Director of the USC – Lockheed Martin Quantum Computing Center. Prior to joining USC, he was the Head of the High-Performance Computing Research Department in the National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory. Prior to joining NERSC, Dr. Lucas was the Deputy Director of DARPA’s Information Technology Office. From 1988 to 1998 he was a member of the research staff of the Institute for Defense Analyses’s Center for Computing Sciences. From 1979 to 1984 he was a member of the Technical Staff of the Hughes Aircraft Company. Dr. Lucas received his BS, MS, and PhD degrees in Electrical Engineering from Stanford University in 1980, 1983, and 1988 respectively.

Posters

Customizable Heterogeneous Architectures

Addressing Architectural Obstacles for Overlay with Stream Network Abstraction, Chengyue Wang, Xiaofan Zhang, Jason Cong, UCLA CS, James Hoe, CMU ECE
Matrix Multiplication by Systolic Array on Versal ACAP, Jason Kimko, Jason Cong, UCLA CS
InTAR: Inter-Task Auto-Reconfigurable Accelerator Design for High Data Volume Variation in DNNs, Zifan He, Huifeng Ke, Jason Cong, UCLA CS
Multi-FPGA Acceleration for Large Scale Conjugate Gradient Solvers, Linghao Song, UCLA CS and Yale Univ.
Understanding Fundamental Challenges of Dataflow Acceleration on General-Purpose Programs, Chris Liu, UCLA CS, Tony Nowatzki, UCLA CS
Leveraging CGRA for Sparse Matrix Multiplication: A Data-Dependent Network Approach, Alice Wu, Tony Nowatzki, UCLA CS
FPGA-Based Hardware Plug-in for Efficient Long-Context Language Model Inference, Jiahao Zhang, UCLA CS

Compilations/Synthesis for Customized Computing

Enola: Compilation for Dynamically Field-Programmable Qubit Arrays with Efficient and Provably Near-Optimal Scheduling, Wan-Hsuan Lin, UCLA CS, Daniel Bochen Tan, UCLA CS & Harvard, Jason Cong, UCLA CS/ECE
Quantum State Preparation Using an Exact CNOT Synthesis Formulation, Hanyu Wang, UCLA CS, Daniel Bochen Tan, UCLA CS & Harvard, Jason Cong, UCLA CS/ECE, Giovanni De Micheli, EPFL CS/ECE
Quantum State Preparation Circuit Optimization Exploiting Don’t Cares, Hanyu Wang, UCLA CS, Daniel Bochen Tan, UCLA CS & Harvard, Jason Cong, UCLA CS/ECE
Enhancing High-Level Synthesis with Automated Pragma Insertion and Code Transformation Framework, Stéphane Pouget, UCLA CS, Louis-Noël Pouchet, Colorado State, Jason Cong, UCLA CS/ECE
Optimizing Latency in HLS through Data-Tile Flow, Stéphane Pouget, UCLA CS, Louis-Noël Pouchet, Colorado State, Jason Cong, UCLA CS/ECE
Unified Optimization of FPGA Dataflow: Task Splitting and Memory Transfers via Nonlinear Programming, Stéphane Pouget, UCLA CS, Louis-Noël Pouchet, Colorado State, Jason Cong, UCLA CS/ECE
StreamHLS: Towards Automatic Dataflow Acceleration, Suhail Basalama, UCLA CS, Jason Cong, UCLA CS/ECE
BenchNoC: Benchmarking the Network-on-Chip on FPGAs, Jake Ke, Sihao Liu, Tony Nowatzki, UCLA CS
OverNoC: Enhancing FPGA Overlay with Network-on-Chip, Sihao Liu, Jake Ke, Tony Nowatzki, UCLA CS
Exploring CGRA Design through Machine Learning, Dylan Kupsh, Tony Nowatzki, UCLA CS

Energy-Efficient ML Models and Novel Applications

HypoSyn: Advancing HLS Modeling with Synthetic Data, Zijian Ding, Tung Nguyen, Weikai Li, Aditya Grover, Yizhou Sun, UCLA CS, Jason Cong, UCLA CS/ECE
LLM-DSE: Chip Design Optimization with an Agentic Workflow, Zijian Ding*, Alice Wu*, Hanyu Wang*, Tony Nowatzki, Yizhou Sun, UCLA CS, Jason Cong, UCLA CS/ECE
Accelerating Large Language Model Pretraining via LFR Pedagogy: Learn, Focus, and Review, Neha Prakriya, Cho-Jui Hsieh, UCLA CS, Jason Cong, UCLA CS/ECE
Knowledge-driven image registration, Yunzheng Zhu and William Hsu, UCLA Radiological Sciences
Vision language model for lung cancer prediction, Luoting Zhuang and William Hsu, UCLA Radiological Sciences
MAS-SAT: Synergizing ML-Assisted and Standalone Solvers for Efficient SAT Solving, Chengdi Cao, Cho-Jui Hsieh, UCLA CS, Jason Cong, UCLA CS/ECE
Knowledge Transfer from FPGA to ASIC HLS, Su Zheng, CUHK/UCLA
Hierarchical Mixture of Experts: Generalizable Learning for High-Level Synthesis, Weikai Li, UCLA CS
Hardware-Informed Machine Learning for High-Level Synthesis, Weikai Li, UCLA CS
An Efficient Rehearsal Scheme for Catastrophic Forgetting Mitigation during Multi-stage Fine-tuning, Andrew Bai, UCLA CS
LoRA Done RITE: Robust Invariant Transformation Equilibration for LoRA Optimization, Jui-Nan Yen, UCLA CS
Dynamic-Width Speculative Beam Decoding for Efficient LLM Inference, Zongyue Qin, UCLA CS

Update the schedule below

Please update the schedule below

There are no upcoming events.

Past Events

CDSC Annual Review – Monday, June 6, 2022 and Tuesday, June 7, 2022
June 6, 2022 – Day 1: 8:00am-8:00pm, Cohen Multipurpose Room 134 Engineering VI (West & East) – No Food & Drink Allowed. June 7, 2022 – Day 2: 8:00am-2:00pm, Cohen Multipurpose Room 134 Engineering... Read More
June 6, 2022 @ 8:00 am - June 7, 2022 @ 2:00 pm
CDSC Annual Review Meeting, February 27-28, 2020
Center for Domain-Specific Computing (CDSC) Annual Review Meeting February 27-28, 2020 Day 1: California Room, UCLA Faculty Center, 480 Charles E Young Dr E, Los Angeles, CA 90095 Day 2: Shannon... Read More
February 27, 2020 @ 8:00 am - February 28, 2020 @ 1:30 pm
CDSC/InTrans Project Annual Review Meeting, February 28-March 1, 2019
CDSC/INTRANS PROJECT ANNUAL REVIEW MEETING FEBRUARY 28 AND MARCH 1, 2019 Mong Learning Center, 180 Auditorium Eng VI (No Food & Drink Allowed) February 28, 2019 Mong Learning Center, 180... Read More
February 28, 2019 - March 1, 2019
CDSC/InTrans Project Annual Review Meeting
CDSC/InTrans Project Annual Review Meeting March 1 and 2, 2018 UCLA Faculty Center, Sequoia Room & Redwood Rooms on Day 1 Mong Learning Center, 180 Auditorium Eng VI on Day... Read More
March 1, 2018 @ 8:00 am - March 2, 2018 @ 1:00 pm
CDSC/InTrans Project Semi-Annual Review Meeting – Day Two
8:30am-9:00am Breakfast 9:00am-10:00am Panel: FPGA vs GPUs (Moderator: Jason Cong) Panelists: Michael Adler (Intel), Kees Vissers (Xilinx), Yuan Xie (UCSB), Zhenman Fang (UCLA 10:00am-11:00am Opportunities of machine learning in healthcare... Read More
June 2, 2017 @ 8:00 am - 1:00 pm