The conference will be held in Shijiazhuang, China from May 26 to 28, 2027, focusing on the frontier field of “Photonic Computing for AI Acceleration” . It will explore key scientific issues and engineering challenges across four major areas, including Integrated Photonic Functional Devices, Photonic Computing Architectures, Models and Training Methods, and System-Level Applications. The conference will delve into cutting-edge technologies such as on-chip integrated photonic devices, optoelectronic heterogeneous integration, diffractive neural networks and free-space computing, optoelectronic hybrid computing architectures, distributed photonic acceleration algorithms for large-scale models, intelligent computational imaging, and quantum-photonic hybrid computing. The conference aims to showcase the latest research achievements through keynote speeches, oral presentations, and poster sessions, promote collaborative innovation between fundamental research and industrial applications, and establish a high-level academic exchange and cooperation platform for global researchers and professionals. We sincerely welcome experts, scholars, technical professionals, and industry representatives from universities, research institutions, enterprises, and related fields worldwide to actively participate and exchange ideas.
Please mark your calendar with the following key deadlines for PCAIA 2027.
All papers related to the theme of Photonic Computing for AI Acceleration can be submitted, including but not limited to the following topics:
On-chip Integrated Photonic Devices
Wavefront Modulation and Metasurface Devices
Optoelectronic Heterogeneous Integration and Packaging
Nonlinear Functional Materials and Devices
On-chip Optical Frequency Combs and Multi-wavelength Light Sources
Diffractive Neural Networks and Free-space Computing
On-chip Integrated Computing Architectures
Optoelectronic Hybrid Computing Architectures
Photonic Computing Architectures with Integrated Storage and Processing
Tensor Parallel Computing and Photonic Operators
Hardware Adaptation and Model Compression for Neural Networks
In-situ Training and Noise-Robust Learning
System Error Calibration and Fault-tolerant Algorithms
Distributed Photonic Acceleration Algorithms for Large-scale Models
Photonic Implementation Schemes for Multimodal Computing
Large-scale Model Inference and Training Acceleration Systems
Intelligent Computational Imaging and Visual Perception
Low-power AI Computing Systems for Edge Devices
Performance Evaluation of Photonic Computing Chips
Quantum-Photonic Hybrid Computing
The full text of each submitted paper will undergo three technical reviews, which include initial review, double-blind peer review (at least three review experts), and final review. The submission will be evaluated on the originality of the content, technical and research content/depth, correctness, relevance to the conference, contribution and readability.
Beijing Zhonghuan Institute of Electric Power and Energy Big Data
International Association For Energy And Environmental Research