ICML 2016 accepted papers on Architectures and efficiency
26 of the 322 papers accepted at ICML 2016 (8.1%) are on Architectures and efficiency. They fall into 5 topics.
See ICML 2016 in the interactive map
Submitting to ICML? 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 · 14 papers
- Deep Speech 2 : End-to-End Speech Recognition in English and Mandarin (2,239 citations)
- Meta-Learning with Memory-Augmented Neural Networks (1,224 citations)
- Unitary Evolution Recurrent Neural Networks (221 citations)
- A Neural Autoregressive Approach to Collaborative Filtering (165 citations)
- Associative Long Short-Term Memory (105 citations)
- Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networks (85 citations)
- Persistent RNNs: Stashing Recurrent Weights On-Chip (67 citations)
- Recurrent Orthogonal Networks and Long-Memory Tasks (55 citations)
- Expressiveness of Rectifier Networks (35 citations)
- Strongly-Typed Recurrent Neural Networks (15 citations)
- Sequence to Sequence Training of CTC-RNNs with Partial Windowing (13 citations)
- Why Regularized Auto-Encoders learn Sparse Representation? (13 citations)
and 2 more in the interactive map.
CNN design · 7 papers
- Group Equivariant Convolutional Networks (552 citations)
- Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units (163 citations)
- Exploiting Cyclic Symmetry in Convolutional Neural Networks (155 citations)
- Multi-Bias Non-linear Activation in Deep Neural Networks (46 citations)
- Convolutional Rectifier Networks as Generalized Tensor Decompositions (21 citations)
- Discrete Deep Feature Extraction: A Theory and New Architectures (10 citations)
- Dynamic Capacity Networks (8 citations)
Attention and transformers · 3 papers
- Ask Me Anything: Dynamic Memory Networks for Natural Language Processing (628 citations)
- A Convolutional Attention Network for Extreme Summarization of Source Code (135 citations)
- Learning to Generate with Memory (14 citations)
Quantization · 1 paper
- Fixed Point Quantization of Deep Convolutional Networks (599 citations)
