Levent Sagun
Artificial Intelligence · 2014–2022
A researcher in the Science corpus: 17 papers; trusts 716 researchers and is trusted by 220 researchers here, as inferred from the citation graph.
Subfields
- Artificial Intelligence — 446 citations into it; #2,418 of 36,961 by trust here, #1,985 by citations
- Computer Vision and Pattern Recognition — 248 citations into it; #7,517 of 21,973 by trust here, #1,989 by citations
- Optimization & Training — 170 citations into it; #819 of 2,605 by trust here, #343 by citations
- Scaling & Learning Theory — 148 citations into it; #153 of 429 by trust here, #120 by citations
- Model Architectures — 80 citations into it
- Representation & Self-Supervised Learning — 71 citations into it
- Statistics and Probability — 66 citations into it; #11,876 of 13,188 by trust here, #1,983 by citations
- Statistical and Nonlinear Physics — 57 citations into it; #3,327 of 10,175 by trust here, #432 by citations
Whom Levent Sagun trusts overall
Their strongest outbound trust, strongest first, inferred from their citations.
- Yann LeCun
- Geoffrey E. Hinton
- Alex Krizhevsky
- L. Eon Bottou
- Yoshua Bengio
- Diederik P. Kingma
- Barak A. Pearlmutter
- Ilya Sutskever
- Kaiming He
- Jimmy Ba
Who trusts Levent Sagun overall
- Vardan Papyan
- Mario Geiger
- Stefano Spigler
- Pratik Chaudhari
- Zi-Mei Yao
- Charles H. Martin
- Zhibin Liao
- Stanislav Fort
- Yoshua Bengio
- Stanislaw Jastrzebski
Papers from Levent Sagun's perspective
Ordered by how strongly the researchers Levent Sagun trusts endorse each paper. Same corpus, a different order for every researcher.
- Deep Residual Learning for Image Recognition (Computer Vision and Pattern Recognition)
- ImageNet classification with deep convolutional neural networks
- Adam: A Method for Stochastic Optimization (International Conference on Learning Representations)
- Long Short-Term Memory
- Gradient-based learning applied to document recognition
- Attention is All you Need (Neural Information Processing Systems)
- ImageNet: A large-scale hierarchical image database
- Very Deep Convolutional Networks for Large-Scale Image Recognition (International Conference on Learning Representations)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift (International Conference on Machine Learning)
- Learning Multiple Layers of Features from Tiny Images
Their papers
- SearchQA: A New Q&A Dataset Augmented with Context from a Search Engine (2017) — 496 citations
- Entropy-SGD: biasing gradient descent into wide valleys (2016) — 882 citations
- ConViT: improving vision transformers with soft convolutional inductive biases (2021) — 1,066 citations
- Empirical Analysis of the Hessian of Over-Parametrized Neural Networks (2017) — 516 citations
- Scaling description of generalization with number of parameters in deep learning (2019) — 214 citations
- Singularity of the Hessian in Deep Learning (2016) — 59 citations
- Eigenvalues of the Hessian in Deep Learning: Singularity and Beyond (2016) — 284 citations
- Finding the Needle in the Haystack with Convolutions: on the benefits of architectural bias (2019) — 37 citations
- Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision (2022) — 125 citations
- A jamming transition from under- to over-parametrization affects loss landscape and generalization (2018) — 59 citations
- A jamming transition from under- to over-parametrization affects generalization in deep learning (2018) — 160 citations
- A Tail-Index Analysis of Stochastic Gradient Noise in Deep Neural Networks (2019) — 354 citations
- Triple descent and the two kinds of overfitting: where and why do they appear? (2020) — 91 citations
- Comparing dynamics: deep neural networks versus glassy systems (2018) — 131 citations
- Energy landscapes for machine learning. (2017) — 101 citations
- The jamming transition as a paradigm to understand the loss landscape of deep neural networks (2018) — 157 citations
- Explorations on high dimensional landscapes (2014) — 64 citations