mlsanityData to 7 Oct 2026

RecLight: A Recurrent Neural Network Accelerator with Integrated Silicon Photonics

arXiv 2209.00084 ↗First posted 31 Aug 2022cs.LG cs.AR cs.NEcited by 12

Not on the map. Why?

Abstract (arXiv)

Recurrent Neural Networks (RNNs) are used in applications that learn dependencies in data sequences, such as speech recognition, human activity recognition, and anomaly detection. In recent years, newer RNN variants, such as GRUs and LSTMs, have been used for implementing these applications. As many of these applications are employed in real-time scenarios, accelerating RNN/LSTM/GRU inference is crucial. In this paper, we propose a novel photonic hardware accelerator called RecLight for accelerating simple RNNs, GRUs, and LSTMs. Simulation results indicate that RecLight achieves 37x lower energy-per-bit and 10% better throughput compared to the state-of-the-art.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 12 of its 37 references that are in the map. First posted on arXiv 31 Aug 2022.

Builds on

12 of 37 references.

References · 12

  1. 2014On the Properties of Neural Machine Translation: Encoder-Decoder Approachescited by 1,303
  2. 2021CrossLight: A Cross-Layer Optimized Silicon Photonic Neural Network Acceleratorcited by 29
  3. 2021A Survey on Silicon Photonics for Deep Learningcited by 23
  4. 2018C-LSTM: Enabling Efficient LSTM using Structured Compression Techniques on FPGAscited by 21
  5. 2020EdgeDRNN: Recurrent Neural Network Accelerator for Edge Inferencecited by 19
  6. 2016Understanding the Impact of Precision Quantization on the Accuracy and Energy of Neural Networkscited by 17
  7. 2021ROBIN: A Robust Optical Binary Neural Network Acceleratorcited by 15
  8. 2021SONIC: A Sparse Neural Network Inference Accelerator with Silicon Photonics for Energy-Efficient Deep Learningcited by 10
  9. 2022A Silicon Photonic Accelerator for Convolutional Neural Networks with Heterogeneous Quantizationcited by 9
  10. 2017Chipmunk: A Systolically Scalable 0.9 mm2, 3.08 Gop/s/mW @ 1.2 mW Accelerator for Near-Sensor Recurrent Neural Network Inferencecited by 6
  11. 2021BPLight-CNN: A Photonics-based Backpropagation Accelerator for Deep Learningcited by 4
  12. 2021ARXON: A Framework for Approximate Communication over Photonic Networks-on-Chipcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 5; 2024: 5; 2025: 1; 2026: 1.

12 papers.

  1. 2023TRON: Transformer Neural Network Acceleration with Non-Coherent Silicon Photonicscited by 12
  2. 2023GHOST: A Graph Neural Network Accelerator using Silicon Photonicscited by 7
  3. 2023Machine Learning Accelerators in 2.5D Chiplet Platforms with Silicon Photonicscited by 5
  4. 2023M3ICRO: Machine Learning-Enabled Compact Photonic Tensor Core based on PRogrammable Multi-Operand Multimode Interferencecited by 5
  5. 2024SafeLight: Enhancing Security in Optical Convolutional Neural Network Acceleratorscited by 4
  6. 2023Cross-Layer Design for AI Acceleration with Non-Coherent Optical Computingcited by 3
  7. 2024Photonics for Sustainable Computingcited by 1
  8. 2024Accelerating Neural Networks for Large Language Models and Graph Processing with Silicon Photonicscited by 1
  9. 2024Silicon Photonic 2.5D Interposer Networks for Overcoming Communication Bottlenecks in Scale-out Machine Learning Hardware Acceleratorscited by 1
  10. 2025PhotoGAN: Generative Adversarial Neural Network Acceleration with Silicon Photonicscited by 1
  11. 2024Optical Computing for Deep Neural Network Acceleration: Foundations, Recent Developments, and Emerging Directionscited by 0
  12. 2026Accelerating Diffusion Models for Generative AI Applications with Silicon Photonicscited by 0