Song Han
Computer Vision and Pattern Recognition · Neural Information Processing Systems · 2015–2024
A researcher in the Science corpus: 48 papers; trusts 3,930 researchers and is trusted by 8,493 researchers here, as inferred from the citation graph.
Subfields
- Computer Vision and Pattern Recognition — 1,403 citations into it; #377 of 21,973 by trust here, #155 by citations
- Artificial Intelligence — 1,255 citations into it; #403 of 36,961 by trust here, #199 by citations
- Efficiency & Compression — 490 citations into it; #4 of 1,273 by trust here, #1 by citations
- Model Architectures — 362 citations into it; #1,584 of 2,288 by trust here, #684 by citations
- Optimization & Training — 192 citations into it; #631 of 2,605 by trust here, #401 by citations
- Evaluation & Benchmarks — 134 citations into it
- Electrical and Electronic Engineering — 133 citations into it; #13,443 of 16,595 by trust here, #2,361 by citations
- Natural Language Processing — 116 citations into it
Whom Song Han trusts overall
Their strongest outbound trust, strongest first, inferred from their citations.
- Alex Krizhevsky
- Kaiming He
- Geoffrey E. Hinton
- Karen Simonyan
- Yann LeCun
- Yoshua Bengio
- Ross B. Girshick
- Ilya Sutskever
- Christian Szegedy
- Diederik P. Kingma
Who trusts Song Han overall
- Ji Lin
- Han Cai
- Yunhe Wang
- Yang He
- Jonathan Frankle
- Chuang Gan
- Xuanyi Dong
- Yi Yang
- Xiaoliang Dai
- Jiahui Yu
Papers from Song Han's perspective
Ordered by how strongly the researchers Song Han trusts endorse each paper. Same corpus, a different order for every researcher.
- ImageNet classification with deep convolutional neural networks
- Deep Residual Learning for Image Recognition (Computer Vision and Pattern Recognition)
- Adam: A Method for Stochastic Optimization (International Conference on Learning Representations)
- ImageNet: A large-scale hierarchical image database
- Long Short-Term Memory
- Attention is All you Need (Neural Information Processing Systems)
- Gradient-based learning applied to document recognition
- Very Deep Convolutional Networks for Large-Scale Image Recognition (International Conference on Learning Representations)
- Going deeper with convolutions
- ImageNet Large Scale Visual Recognition Challenge
Their papers
- Deep Compression: Compressing Deep Neural Network with Pruning, Trained Quantization and Huffman Coding (2015) — 10,226 citations
- Learning both Weights and Connections for Efficient Neural Network (2015) — 7,826 citations
- ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware (2018) — 2,104 citations
- SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size (2016) — 8,474 citations
- TSM: Temporal Shift Module for Efficient Video Understanding (2018) — 2,128 citations
- Trained Ternary Quantization (2016) — 1,089 citations
- AMC: AutoML for Model Compression and Acceleration on Mobile Devices (2018) — 1,502 citations
- Once for All: Train One Network and Specialize it for Efficient Deployment (2019) — 1,586 citations
- EIE: Efficient Inference Engine on Compressed Deep Neural Network (2016) — 2,611 citations
- Differentiable Augmentation for Data-Efficient GAN Training (2020) — 694 citations
- HAQ: Hardware-Aware Automated Quantization With Mixed Precision (2018) — 1,120 citations
- Point-Voxel CNN for Efficient 3D Deep Learning (2019) — 855 citations
- Temporal Shift Module for Efficient Video Understanding (2018) — 221 citations
- DSD: Regularizing Deep Neural Networks with Dense-Sparse-Dense Training Flow (2016) — 83 citations
- Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training (2017) — 1,676 citations
- Lite Transformer with Long-Short Range Attention (2020) — 392 citations
- Path-Level Network Transformation for Efficient Architecture Search (2018) — 221 citations
- - LEVEL ACCURACY WITH 50 X FEWER PARAMETERS AND < 0 . 5 MB MODEL SIZE (2016)
- SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models (2022) — 1,734 citations
- HAT: Hardware-Aware Transformers for Efficient Natural Language Processing (2020) — 292 citations