ZJZiheng Jiang

Ziheng Jiang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Ziheng Jiang has a total of 22 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • High Accuracy Near-Field Electromagnetic Emission Identification System Using Characteristics Recognition and Image Localization

    by Yanhua Peng, Hui Xu, Hangyi Zhou, Ziheng Jiang, Dongdong Shen, Shuo Cui, Donglin Su

    In this paper, an automatic near-field measurement system is introduced to record the electromagnetic emission (EME) of equipment under testing (EUT). Firstly, a camera is used to capture the measurement area, which is combined with an image positioning method to achieve fast and accurate positioning of the area. Then, based on the wavelet transform and autocorrelation transform of the EME signal identification algorithm, the system further identifies the broadband and narrowband characteristics of the signal and obtains the power distributions of the corresponding characteristics. In performance tests, the near-field probe moves with a minimum stepping accuracy of 0.1 mm. The movement error of the consistency and stability are guaranteed as less than 1% of the set value. The image positioning error in the 30 cm × 30 cm scanned area is less than 0.5 mm. Finally, the system identifies EME signals from a circuit with multiple crystal oscillators and a sine signal, predictive identification of the circuit functions under the metal shell of a router system. The proposed measurement system therefore offers great measurement accuracy, stability, and efficiency.


  • Open Access

    Relay: A High-Level Compiler for Deep Learning

    by Jared Roesch, Steven Lyubomirsky, Marisa Kirisame, Logan Weber, Josh Pollock, Luis Vega, Ziheng Jiang, Tianqi Chen, Thierry Moreau, Zachary Tatlock

    Frameworks for writing, compiling, and optimizing deep learning (DL) models have recently enabled progress in areas like computer vision and natural language processing. Extending these frameworks to accommodate the rapidly diversifying landscape of DL models and hardware platforms presents challenging tradeoffs between expressivity, composability, and portability. We present Relay, a new compiler framework for DL. Relay's functional, statically typed intermediate representation (IR) unifies and generalizes existing DL IRs to express state-of-the-art models. The introduction of Relay's expressive IR requires careful design of domain-specific optimizations, addressed via Relay's extension mechanisms. Using these extension mechanisms, Relay supports a unified compiler that can target a variety of hardware platforms. Our evaluation demonstrates Relay's competitive performance for a broad class of models and devices (CPUs, GPUs, and emerging accelerators). Relay's design demonstrates how a unified IR can provide expressivity, composability, and portability without compromising performance.

  • Open Access

    A Hardware-Software Blueprint for Flexible Deep Learning Specialization

    by Thierry Moreau, Tianqi Chen, Luis Vega, Jared Roesch, Eddie Yan, Lianmin Zheng, Josh Fromm, Ziheng Jiang, Luis Ceze, Carlos Guestrin, et al.

    Specialized Deep Learning (DL) acceleration stacks, designed for a specific set of frameworks, model architectures, operators, and data types, offer the allure of high performance while sacrificing flexibility. Changes in algorithms, models, operators, or numerical systems threaten the viability of specialized hardware accelerators. We propose VTA, a programmable deep learning architecture template designed to be extensible in the face of evolving workloads. VTA achieves this flexibility via a parametrizable architecture, two-level ISA, and a JIT compiler. The two-level ISA is based on (1) a task-ISA that explicitly orchestrates concurrent compute and memory tasks and (2) a microcode-ISA which implements a wide variety of operators with single-cycle tensor-tensor operations. Next, we propose a runtime system equipped with a JIT compiler for flexible code-generation and heterogeneous execution that enables effective use of the VTA architecture. VTA is integrated and open-sourced into Apache TVM, a state-of-the-art deep learning compilation stack that provides flexibility for diverse models and divergent hardware backends. We propose a flow that performs design space exploration to generate a customized hardware architecture and software operator library that can be leveraged by mainstream learning frameworks. We demonstrate our approach by deploying optimized deep learning models used for object classification and style transfer on edge-class FPGAs.