JMLR 2021 papers on Architectures and efficiency
16 of the 285 papers published in JMLR 2021 (5.6%) are on Architectures and efficiency, compared with 1.6% at JMLR 2020. They fall into 6 topics.
See JMLR 2021 in the interactive map
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 · 9 papers
- LassoNet: A Neural Network with Feature Sparsity (107 citations)
- Kernel Operations on the GPU, with Autodiff, without Memory Overflows (99 citations)
- Phase Diagram for Two-layer ReLU Neural Networks at Infinite-width Limit (20 citations)
- A Greedy Algorithm for Quantizing Neural Networks (13 citations)
- Understanding Recurrent Neural Networks Using Nonequilibrium Response Theory (11 citations)
- Further results on latent discourse models and word embeddings (1 citation)
- TensorHive: Management of Exclusive GPU Access for Distributed Machine Learning Workloads (1 citation)
- Doubly infinite residual neural networks: a diffusion process approach
- Implicit Self-Regularization in Deep Neural Networks: Evidence from Random Matrix Theory and Implications for Learning
Pruning and efficient tuning · 2 papers
Quantization · 2 papers
- CAT: Compression-Aware Training for bandwidth reduction (11 citations)
- sklvq: Scikit Learning Vector Quantization (11 citations)
Attention and transformers · 1 paper
- Attention is Turing-Complete (20 citations)
CNN design · 1 paper
Spiking networks · 1 paper
- Bifurcation Spiking Neural Network (2 citations)
