NeurIPS 2016 accepted papers on Architectures and efficiency
42 of the 570 papers accepted at NeurIPS 2016 (7.4%) are on Architectures and efficiency. They fall into 5 topics.
See NeurIPS 2016 in the interactive map
Submitting to NeurIPS? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
RNNs and normalization · 23 papers
- Deep ADMM-Net for Compressive Sensing MRI (933 citations)
- Binarized Neural Networks (907 citations)
- Residual Networks Behave Like Ensembles of Relatively Shallow Networks (596 citations)
- Direct Feedback Alignment Provides Learning in Deep Neural Networks (294 citations)
- RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism (220 citations)
- Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences (211 citations)
- Learning the Number of Neurons in Deep Networks (187 citations)
- Swapout: Learning an ensemble of deep architectures (121 citations)
- Maximal Sparsity with Deep Networks? (113 citations)
- Architectural Complexity Measures of Recurrent Neural Networks (110 citations)
- Full-Capacity Unitary Recurrent Neural Networks (104 citations)
- On Multiplicative Integration with Recurrent Neural Networks (58 citations)
and 11 more in the interactive map.
CNN design · 14 papers
- Dynamic Filter Networks (474 citations)
- Learning Structured Sparsity in Deep Neural Networks (458 citations)
- Deep Learning Models of the Retinal Response to Natural Scenes (186 citations)
- Convolutional Neural Fabrics (170 citations)
- Learning shape correspondence with anisotropic convolutional neural networks (169 citations)
- CNNpack: Packing Convolutional Neural Networks in the Frequency Domain (154 citations)
- PerforatedCNNs: Acceleration through Elimination of Redundant Convolutions (117 citations)
- Incremental Boosting Convolutional Neural Network for Facial Action Unit Recognition (89 citations)
- CRF-CNN: Modeling Structured Information in Human Pose Estimation (37 citations)
- Doubly Convolutional Neural Networks (32 citations)
- Learning Sensor Multiplexing Design through Back-propagation (28 citations)
- End-to-End Kernel Learning with Supervised Convolutional Kernel Networks (7 citations)
and 2 more in the interactive map.
Attention and transformers · 3 papers
- An Online Sequence-to-Sequence Model Using Partial Conditioning (53 citations)
- Latent Attention For If-Then Program Synthesis (46 citations)
- Can Active Memory Replace Attention? (16 citations)
Pruning and efficient tuning · 1 paper
- Dynamic Network Surgery for Efficient DNNs (140 citations)
