M. Smelyanskiy

Artificial Intelligence · 2010–2021

A researcher in the Science corpus: 14 papers; trusts 83 researchers and is trusted by 40 researchers here, as inferred from the citation graph.

Subfields

• Artificial Intelligence — 217 citations into it; #5,904 of 35,345 by trust here, #1,659 by citations
• Computer Vision and Pattern Recognition — 193 citations into it
• Optimization & Training — 77 citations into it; #796 of 2,601 by trust here, #210 by citations
• Hardware and Architecture — 59 citations into it
• Computer Networks and Communications — 42 citations into it
• Model Architectures — 41 citations into it
• Natural Language Processing — 34 citations into it
• Management Science and Operations Research — 29 citations into it

Whom M. Smelyanskiy trusts overall

Their strongest outbound trust, strongest first, inferred from their citations.

1. Geoffrey E. Hinton
2. L. Eon Bottou
3. Alex Krizhevsky
4. Kaiming He
5. Yann LeCun
6. Diederik P. Kingma
7. Karen Simonyan
8. Alex Graves
9. Ilya Sutskever
10. Yoshua Bengio

Who trusts M. Smelyanskiy overall

1. Behnam Neyshabur
2. Yang You
3. Stanislav Fort
4. Stanislaw Jastrzebski
5. Nicholas Carlini
6. Yuanzhi Li

Papers from M. Smelyanskiy's perspective

Ordered by how strongly the researchers M. Smelyanskiy trusts endorse each paper. Same corpus, a different order for every researcher.

1. ImageNet classification with deep convolutional neural networks
2. Deep Residual Learning for Image Recognition (Computer Vision and Pattern Recognition)
3. Long Short-Term Memory
4. Attention is All you Need (Neural Information Processing Systems)
5. ImageNet: A large-scale hierarchical image database
6. Very Deep Convolutional Networks for Large-Scale Image Recognition (International Conference on Learning Representations)
7. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift (International Conference on Machine Learning)
8. Gradient-based learning applied to document recognition
9. ImageNet Large Scale Visual Recognition Challenge
10. Going deeper with convolutions

Their papers

• On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima (2016) — 3,531 citations
• A Study of BFLOAT16 for Deep Learning Training (2019) — 492 citations
• Glow: Graph Lowering Compiler Techniques for Neural Networks (2018) — 337 citations
• Applied Machine Learning at Facebook: A Datacenter Infrastructure Perspective (2018) — 626 citations
• Deep Learning Training in Facebook Data Centers: Design of Scale-up and Scale-out Systems (2020) — 107 citations
• FBGEMM: Enabling High-Performance Low-Precision Deep Learning Inference (2021) — 69 citations
• Deep Learning Recommendation Model for Personalization and Recommendation Systems (2019) — 944 citations
• Low-Precision Hardware Architectures Meet Recommendation Model Inference at Scale (2021) — 14 citations
• Software-hardware co-design for fast and scalable training of deep learning recommendation models (2021) — 209 citations
• High-performance, Distributed Training of Large-scale Deep Learning Recommendation Models (2021) — 49 citations
• RecNMP: Accelerating Personalized Recommendation with Near-Memory Processing (2019) — 273 citations
• The Architectural Implications of Facebook's DNN-Based Personalized Recommendation (2019) — 330 citations
• Deep Learning Inference in Facebook Data Centers: Characterization, Performance Optimizations and Hardware Implications (2018) — 205 citations
• Debunking the 100X GPU vs. CPU myth: an evaluation of throughput computing on CPU and GPU (2010) — 854 citations