Convolutional networks
Compute interpretation
Architecture family that exploits locality and weight sharing for GPU-friendly dense vision workloads.
Supporting reading cards
- Gradient-based learning applied to document recognition (1998,
pre_2012_cpu_statistical_foundations) - ImageNet: A large-scale hierarchical image database (2009,
pre_2012_cpu_statistical_foundations) - ImageNet Classification with Deep Convolutional Neural Networks (2012,
single_gpu_deep_learning) - Rich feature hierarchies for accurate object detection and semantic segmentation (2014,
single_gpu_deep_learning) - Very Deep Convolutional Networks for Large-Scale Image Recognition (2014,
single_gpu_deep_learning) - Going Deeper with Convolutions (2014,
single_gpu_deep_learning) - Fast R-CNN (2015,
single_gpu_deep_learning) - Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (2015,
single_gpu_deep_learning) - U-Net: Convolutional Networks for Biomedical Image Segmentation (2015,
single_gpu_deep_learning) - Deep Residual Learning for Image Recognition (2015,
multi_gpu_dense_training) - Rethinking the Inception Architecture for Computer Vision (2015,
multi_gpu_dense_training) - Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks (2015,
generative_media_compute) - Identity Mappings in Deep Residual Networks (2016,
multi_gpu_dense_training) - Xception: Deep Learning with Depthwise Separable Convolutions (2016,
multi_gpu_dense_training) - Densely Connected Convolutional Networks (2016,
multi_gpu_dense_training) - Image-to-Image Translation with Conditional Adversarial Networks (2016,
generative_media_compute) - SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size (2016,
efficient_edge_inference) - Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks (2017,
generative_media_compute) - MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications (2017,
efficient_edge_inference) - EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks (2019,
multi_gpu_dense_training) - An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (2020,
tpu_accelerator_transformer_era)
Obsolete or less central under later compute
Track this only through linked reading cards; do not treat this method page as standalone evidence.