About

I am currently an assistant professor in the School of Electrical, Computer and Energy Engineering at Arizona State University.

  • CV
  • Please visit our Lab ScopeX homepage for more information and openings

I received my Ph.D. degree at the Department of Electrical and Computer Engineering, The University of Texas at Austin, in 2023, advisored by Prof. David Z. Pan and co-advised by Prof. Ray T. Chen, and received the B.E. degree in microelectronic science and engineering from Fudan University, in 2018.

My current research interests include

  • Emerging Hardware Design for Efficient Computing
    • Efficient AI hardware design
    • Electronic-photonic mixed-signal platform for high-performance, energy-efficient computing
  • Efficient Algorithm, Co-Design & Automation
    • Hardware-algorithm co-design for emerging hardware (photonics, post-CMOS electronics, quantum)
    • Efficient ML model/algorithm
    • Design automation for electronic-photonic heterogeneous platforms
  • Parallel Computing and GPU Acceleration for VLSI Physical Design Automation

I have received the Best Paper Award at ASP-DAC 2020, the Best Paper Finalist at DAC 2020, the Best Poster Award at NSF Workshop on Machine Learning Hardware (2020), and the Best Paper Award at IEEE TCAD 2021. I have won the ACM/SIGDA Student Research Competition First Place (2020), the ACM Student Research Competition Grand Finals First Place (2021), the Robert S. Hilbert Memorial Optical Design Competition 2022, and Margarida Jacome Dissertation Prize at UT Austin ECE (2023).

News

  • 03/2024: One co-authored paper, FPQA-C: A Compilation Framework for Field Programmable Qubit Array, is accepted by ISCA 2024. Cheers!

  • 02/2024: One co-authored paper, Q-Pilot: Field Programmable Quantum Array Compilation with Flying Ancillas, is accepted by DAC 2024. Cheers!

  • 01/2024: Our recent work, M3ICRO: Machine Learning-Enabled Compact Photonic Tensor Core based on PRogrammable Multi-Operand Multimode Interference, is accepted by APL Machine Learning 2024. Cheers!

  • 12/2023: One co-authored paper, Integrated Multi-Operand Optical Neurons for Scalable and Hardware-Efficient Deep Learning, is accepted by Nanophotonics 2023. Cheers!

  • 10/2023: One co-authored paper, Lightening-Transformer: A Dynamically-operated Optically-interconnected Photonic Transformer Accelerator, is accepted by HPCA 2023. Cheers!

  • 09/2023: One co-authored paper, Pre-RMSNorm and Pre-CRMSNorm Transformers: Equivalent and Efficient Pre-LN Transformers, is accepted by NeurIPS 2023 and selected as Spotlight paper. Cheers!

  • 09/2023: Our recent work, Domain-Specific Optimization for Quantized Optical AI Computing Systems, is accepted by ICCV LBQNN workshop 2023. Cheers!

  • 07/2023: We have organized a tutorial at DAC 2023: A Journey to Optical Computing: From Physics Fundamentals to Hardware-Software Co-Design, Automation, and Application. Please check out our slides.

  • 06/2023: Our recent work, DOTA: A Dynamically-Operated Photonic Tensor Core for Energy-Efficient Transformer Accelerator, is accepted to MLSys SNAP 2023. Cheers!

  • 11/2022: Our recent work, A compact butterfly-style silicon photonic-electronic neural chip for hardware-efficient deep learning, is accepted by ACS Photonics. Cheers!

  • 10/2022: Our recent work, HEAT: Hardware-Efficient Automatic Tensor Decomposition for Transformer Compression, is accepted by NeurIPS MLSys Workshop 2022 as a Spotlight paper. Cheers!

  • 09/2022: Our recent work, NeurOLight: A Physics-Agnostic Neural Operator Enabling Parametric Photonic Device Simulation, is accepted by NeurIPS 2022 as a Spotlight paper. Cheers!

  • 07/2022: Prof. David Pan and I will give a talk at ACCESS-CEDA seminar, Light-AI Interaction: The Convergence of Photonic Deep Learning and Cross-Layer Design Automation.

  • 07/2022: Our research project, A compact 4x4 butterfly-style silicon photonic-electronic neural chip for hardware-efficient deep learning, won the Robert S. Hilbert Memorial Optical Design Competition 2022. Cheers!

  • 07/2022: One co-authored paper, Fuse and Mix: MACAM-Enabled Analog Activation for Energy-Efficient Neural Acceleration, is accepted by IEEE ICCAD 2022. Cheers!

  • 07/2022: Our recent work, SqueezeLight: A Multi-Operand Ring-Based Optical Neural Network with Cross-Layer Scalability, is accepted by IEEE TCAD 2022. Cheers!

  • 05/2022: One co-authored paper, ELight: Towards Efficient and Aging-Resilient Photonic In-Memory Neurocomputing, is accepted by IEEE TCAD 2022. Cheers!

  • 03/2022: Our recent work, Multi-Scale High-Resolution Vision Transformer for Semantic Segmentation, is accepted by CVPR 2022. Cheers!

  • 02/2022: Our recent work, ADEPT: Automatic Differentiable DEsign of Photonic Tensor Cores, is accepted by DAC 2022. Cheers!

  • 02/2022: One co-authored paper, On-Chip QNN: Towards Efficient On-Chip Training of Quantum Neural Networks, is accepted by DAC 2022. Cheers!

  • 02/2022: One co-authored paper, RobustQNN: Noise-Aware Training for Robust Quantum Neural Networks, is accepted by DAC 2022. Cheers!

  • 02/2022: One co-authored paper, A Timing Engine Inspired Graph Neural Network Model for Pre-Routing Slack Prediction, is accepted by DAC 2022. Cheers!

  • 10/2021: One co-authored paper, QuantumNAS: Noise-Adaptive Search for Robust Quantum Circuits, is accepted by HPCA 2022. Cheers!

  • 10/2021: One co-authored paper, DREAMPlace: Deep Learning Toolkit-Enabled GPU Acceleration for Modern VLSI Placement, received the Donald O. Pederson Best Paper Award at IEEE TCAD 2021. Cheers! Please check the UT ECE News.

  • 09/2021: Our recent work, L2ight: Enabling On-Chip Learning for Optical Neural Networks via Efficient in-situ Subspace Optimization, is accepted by NeurIPS 2021. Cheers!

  • 09/2021: One co-authored paper, ELight: Enabling Efficient Photonic In-Memory Neurocomputing with Life Enhancement, is accepted by ASP-DAC 2022. Cheers!

  • 09/2021: We are pleased to announce DREAMPlace 3.0 with features to support region constraints with multi-electrostatic fields and enhanced gradient descent optimization. Please check it out on Github. Cheers!

  • 07/2021: Our recent work, Towards Memory-Efficient Neural Networks via Multi-Level in situ Generation, is accepted by ICCV 2021. Cheers!

  • 06/2021: A PyTorch-centric Optical Neural Network library pytorch-onn that supports fast development and training for ONNs is released!

  • 05/2021: Our research project Light in Artificial Intelligence: Efficient Neurocomputing with Optical Neural Networks won the First Place in ACM Student Research Competition (SRC) 2021 Grand Finals. Cheers! Please check ACM SRC website and find our work at this link.

  • 05/2021: One co-authored paper, Towards high-speed and energy-efficient computing: A WDM-based scalable on-chip silicon integrated optical comparator, is accepted by Laser & Photonics Reviews (IF: 9.6). Cheers!

  • 03/2021: One co-authored paper, Optimizer Fusion: Efficient Training with Better Locality and Parallelism, is accepted by ICLR Workshop, HAET 2021. Cheers!

  • 02/2021: We release an open-source PyTorch-centric hybrid classical-quantum machine learning framework pytorch-quantum to support fast and easy development of quantum machine learning models.

  • 12/2020: Our work, FLOPS: Efficient On-Chip Learning for Optical Neural Networks Through Stochastic Zeroth-Order Optimization, won the Best Poster Award at NSF Workshop on Machine Learning Hardware 2020. Cheers!

  • 12/2020: Our recent work, Efficient On-Chip Learning for Optical Neural Networks Through Power-Aware Sparse Zeroth-Order Optimization, is accepted by AAAI 2021. Cheers!

  • 11/2020: Our recent work, SqueezeLight: Towards Scalable Optical Neural Networks with Multi-Operand Ring Resonators, is accepted by DATE 2021. Cheers!

  • 11/2020: Our recent work, O2NN: Optical Neural Networks with Differential Detection-Enabled Optical Operands, is accepted by DATE 2021. Cheers!

  • 11/2020: Our research project Light in Artificial Intelligence: Efficient Neuromorphic Computing with Optical Neural Networks won the First Place in ACM/SIGDA Student Research Competition 2020. Cheers! Please check the UT ECE News and ACM SRC website!

  • 10/2020: One co-authored paper, Sequential logic and pipelining in chip-based electronic-photonic digital computing, is accepted by IEEE Photonics Journal. Cheers!

  • 09/2020: Our recent work, Towards Hardware-Efficient Optical Neural Networks: Beyond FFT Architecture via Joint Learnability, is accepted by IEEE TCAD. Cheers!

  • 08/2020: One co-authored paper, Wavelength-division-multiplexing (WDM)-based integrated electronic–photonic switching network (EPSN) for high-speed data processing and transportation, is accepted by Nanophotonics (IF: 7.5). Cheers!

  • 07/2020: Our recent work, FLOPS: Efficient On-Chip Learning for Optical Neural Networks Through Stochastic Zeroth-Order Optimization, was selected as one out of 6 Best Paper Finalists at DAC 2020. Cheers!

  • 07/2020: Our recent work, DREAMPlace 3.0: Multi-Electrostatics Based Robust VLSI Placement with Region Constraints, is accepted by ICCAD 2020. Cheers!

  • 07/2020: One co-authored paper, An Efficient Training Framework for Reversible Neural Architectures, is accepted by ECCV 2020. Cheers!

  • 06/2020: One co-authored paper, DREAMPlace: Deep Learning Toolkit-Enabled GPU Acceleration for Modern VLSI Placement, is accepted by IEEE TCAD. Please checkout the PDF and source code release. Cheers!

  • 06/2020: Our recent work, FLOPS: Efficient On-Chip Learning for Optical Neural Networks Through Stochastic Zeroth-Order Optimization, is accepted by DAC 2020. Cheers!

  • 04/2020: One co-authored paper, Electronic-photonic Arithmetic Logic Unit for High-speed Computing, is accepted by Nature Communications (IF: 13.8). Please checkout the PDF. Cheers!

  • 03/2020: Our recent work, ROQ: A Noise-Aware Quantization Scheme Towards Robust Optical Neural Networks with Low-bit Controls, is accepted by DATE 2020. Please checkout the PDF. Cheers!

  • 02/2020: One co-authored paper, ABCDPlace: Accelerated Batch-based Concurrent Detailed Placement on Multi-threaded CPUs and GPUs, is accepted by IEEE TCAD. Please checkout the PDF and source code release. Cheers!

  • 01/2020: Our recent work, Towards Area-Efficient Optical Neural Networks: An FFT-based Architecture, received the Best Paper Award at ASP-DAC 2020. Please checkout the PDF. Cheers and thanks to all the co-authors! Please check the UT ECE News!

  • 11/2019: One co-authored paper, Design Technology for Scalable and Robust Photonic Integrated Circuits, is accepted by ICCAD 2019. Please checkout the PDF. Cheers!