mlsanityData to 14 Oct 2026

Efficiera Residual Networks: Hardware-Friendly Fully Binary Weight with 2-bit Activation Model Achieves Practical ImageNet Accuracy

arXiv 2410.11553 ↗First posted 15 Oct 2024cs.CV cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

The edge-device environment imposes severe resource limitations, encompassing computation costs, hardware resource usage, and energy consumption for deploying deep neural network models. Ultra-low-bit quantization and hardware accelerators have been explored as promising approaches to address these challenges. Ultra-low-bit quantization significantly reduces the model size and the computational cost. Despite progress so far, many competitive ultra-low-bit models still partially rely on float or non-ultra-low-bit quantized computation such as the input and output layer. We introduce Efficiera Residual Networks (ERNs), a model optimized for low-resource edge devices. ERNs achieve full ultra-low-bit quantization, with all weights, including the initial and output layers, being binary, and activations set at 2 bits. We introduce the shared constant scaling factor technique to enable integer-valued computation in residual connections, allowing our model to operate without float values until the final convolution layer. Demonstrating competitiveness, ERNs achieve an ImageNet top-1 accuracy of 72.5pt with a ResNet50-compatible architecture and 63.6pt with a model size less than 1MB. Moreover, ERNs exhibit impressive inference times, reaching 300FPS with the smallest model and 60FPS with the largest model on a cost-efficient FPGA device.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 24 of its 40 references that are in the map. First posted on arXiv 15 Oct 2024.

Builds on

24 of 40 references.

References · 24

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  3. 2017Squeeze-and-Excitation Networkscited by 4,789
  4. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,269
  5. 2020Designing Network Design Spacescited by 960
  6. 2016XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networkscited by 849
  7. 2019Learned Step Size Quantizationcited by 628
  8. 2017In-Datacenter Performance Analysis of a Tensor Processing Unitcited by 618
  9. 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 569
  10. 2018LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networkscited by 274
  11. 2018Bi-Real Net: Enhancing the Performance of 1-bit CNNs With Improved Representational Capability and Advanced Training Algorithmcited by 235
  12. 2019Fixing the train-test resolution discrepancycited by 209
12 more
  1. 2020ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functionscited by 184
  2. 2017Deep Learning with Low Precision by Half-wave Gaussian Quantizationcited by 117
  3. 2021Network Quantization with Element-wise Gradient Scalingcited by 87
  4. 2021PokeBNN: A Binary Pursuit of Lightweight Accuracycited by 49
  5. 2022AdaBin: Improving Binary Neural Networks with Adaptive Binary Setscited by 38
  6. 2020FracBNN: Accurate and FPGA-Efficient Binary Neural Networks with Fractional Activationscited by 32
  7. 2019daBNN: A Super Fast Inference Framework for Binary Neural Networks on ARM devicescited by 27
  8. 2018Synetgy: Algorithm-hardware Co-design for ConvNet Accelerators on Embedded FPGAscited by 18
  9. 2018Packing Sparse Convolutional Neural Networks for Efficient Systolic Array Implementations: Column Combining Under Joint Optimizationcited by 15
  10. 2022Join the High Accuracy Club on ImageNet with A Binary Neural Network Ticketcited by 14
  11. 2018Binary Input Layer: Training of CNN models with binary input datacited by 4
  12. 2024Pixel Embedding: Fully Quantized Convolutional Neural Network with Differentiable Lookup Tablecited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023ULEEN: A Novel Architecture for Ultra Low-Energy Edge Neural Networks
  2. 2025Batch Normalization-Free Fully Integer Quantized Neural Networks via Progressive Tandem Learning
  3. 20213U-EdgeAI: Ultra-Low Memory Training, Ultra-Low BitwidthQuantization, and Ultra-Low Latency Acceleration
  4. 2025FF-INT8: Efficient Forward-Forward DNN Training on Edge Devices with INT8 Precision
  5. 2026Characterizing the Impact of NVFP4 Quantization for Low-Power Edge AI Deployment
  6. 2021Enabling Binary Neural Network Training on the Edge
  7. 2025Low-bit Model Quantization for Deep Neural Networks: A Survey
  8. 2025EfficientQuant: An Efficient Post-Training Quantization for CNN-Transformer Hybrid Models on Edge Devices