mlsanityData to 7 Oct 2026

PrivCirNet: Efficient Private Inference via Block Circulant Transformation

arXiv 2405.14569 ↗First posted 23 May 2024cs.CR cs.AIcited by 7

Not on the map. Why?

Abstract (arXiv)

Homomorphic encryption (HE)-based deep neural network (DNN) inference protects data and model privacy but suffers from significant computation overhead. We observe transforming the DNN weights into circulant matrices converts general matrix-vector multiplications into HE-friendly 1-dimensional convolutions, drastically reducing the HE computation cost. Hence, in this paper, we propose method, a protocol/network co-optimization framework based on block circulant transformation. At the protocol level, PrivCirNet customizes the HE encoding algorithm that is fully compatible with the block circulant transformation and reduces the computation latency in proportion to the block size. At the network level, we propose a latency-aware formulation to search for the layer-wise block size assignment based on second-order information. PrivCirNet also leverages layer fusion to further reduce the inference cost. We compare PrivCirNet with the state-of-the-art HE-based framework Bolt (IEEE S&P 2024) and the HE-friendly pruning method SpENCNN (ICML 2023). For ResNet-18 and Vision Transformer (ViT) on Tiny ImageNet, PrivCirNet reduces latency by 5.0× and 1.3× with iso-accuracy over Bolt, respectively, and improves accuracy by 4.1% and 12% over SpENCNN, respectively. For MobileNetV2 on ImageNet, PrivCirNet achieves 1.7× lower latency and 4.2% better accuracy over Bolt and SpENCNN, respectively. Our code and checkpoints are available on Git Hub.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 18 of its 58 references that are in the map. First posted on arXiv 23 May 2024.

Builds on

18 of 58 references.

References · 18

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 464
  3. 2021Escaping the Big Data Paradigm with Compact Transformerscited by 261
  4. 2019HAWQ-V2: Hessian Aware trace-Weighted Quantization of Neural Networkscited by 250
  5. 2022MPCFormer: fast, performant and private Transformer inference with MPCcited by 67
  6. 2021DeepReDuce: ReLU Reduction for Fast Private Inferencecited by 50
  7. 2019SHE: A Fast and Accurate Deep Neural Network for Encrypted Datacited by 36
  8. 2022Selective Network Linearization for Efficient Private Inferencecited by 31
  9. 2023Learning to Linearize Deep Neural Networks for Secure and Efficient Private Inferencecited by 20
  10. 2022MPCViT: Searching for Accurate and Efficient MPC-Friendly Vision Transformer with Heterogeneous Attentioncited by 19
  11. 2021GALA: Greedy ComputAtion for Linear Algebra in Privacy-Preserved Neural Networkscited by 10
  12. 2023HyPHEN: A Hybrid Packing Method and Optimizations for Homomorphic Encryption-Based Neural Networkscited by 9
6 more
  1. 2023CoPriv: Network/Protocol Co-Optimization for Communication-Efficient Private Inferencecited by 6
  2. 2023NeuJeans: Private Neural Network Inference with Joint Optimization of Convolution and FHE Bootstrappingcited by 6
  3. 2023Falcon: Accelerating Homomorphically Encrypted Convolutions for Efficient Private Mobile Network Inferencecited by 5
  4. 2022Impala: Low-Latency, Communication-Efficient Private Deep Learning Inferencecited by 3
  5. 2023Toward Practical Privacy-Preserving Convolutional Neural Networks Exploiting Fully Homomorphic Encryptioncited by 3
  6. 2022Efficient Pruning for Machine Learning Under Homomorphic Encryptioncited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 4.

7 papers.

  1. 2025CipherPrune: Efficient and Scalable Private Transformer Inferencecited by 9
  2. 2025Privacy-Preserving Semantic Segmentation from Ultra-Low-Resolution RGB Inputscited by 1
  3. 2025An Efficient Private GPT Never Autoregressively Decodescited by 0
  4. 2026PRISM: Sensitivity-Aware PolynoMial PRuning for EffIcient Neural Network Encryptioncited by 0
  5. 2026Gecko: Fast Private Inference via Secure Public Encoder Offloadingcited by 0
  6. 2026OptiPrime: Optimizing Private Inference through Protocol-Hardware Co-designcited by 0
  7. 2026Fiona: Accelerating FHE Inference with Packing-Aware Ternary Weightscited by 0