Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Stanislav Fort

How many degrees of freedom do we need to train deep networks: a loss landscape perspective

Jul 13, 2021
Brett W. Larsen, Stanislav Fort, Nic Becker, Surya Ganguli

  Access Paper or Ask Questions

A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detection

Jun 16, 2021
Jie Ren, Stanislav Fort, Jeremiah Liu, Abhijit Guha Roy, Shreyas Padhy, Balaji Lakshminarayanan

  Access Paper or Ask Questions

Exploring the Limits of Out-of-Distribution Detection

Jun 06, 2021
Stanislav Fort, Jie Ren, Balaji Lakshminarayanan

* S.F. and J.R. contributed equally 

  Access Paper or Ask Questions

Drawing Multiple Augmentation Samples Per Image During Training Efficiently Decreases Test Error

May 27, 2021
Stanislav Fort, Andrew Brock, Razvan Pascanu, Soham De, Samuel L. Smith

  Access Paper or Ask Questions

Analyzing Monotonic Linear Interpolation in Neural Network Loss Landscapes

Apr 23, 2021
James Lucas, Juhan Bae, Michael R. Zhang, Stanislav Fort, Richard Zemel, Roger Grosse

* 15 pages in main paper, 4 pages of references, 24 pages in appendix. 29 figures in total 

  Access Paper or Ask Questions

Deep learning versus kernel learning: an empirical study of loss landscape geometry and the time evolution of the Neural Tangent Kernel

Oct 28, 2020
Stanislav Fort, Gintare Karolina Dziugaite, Mansheej Paul, Sepideh Kharaghani, Daniel M. Roy, Surya Ganguli

* 19 pages, 19 figures, In Advances in Neural Information Processing Systems 34 (NeurIPS 2020) 

  Access Paper or Ask Questions

Training independent subnetworks for robust prediction

Oct 13, 2020
Marton Havasi, Rodolphe Jenatton, Stanislav Fort, Jeremiah Zhe Liu, Jasper Snoek, Balaji Lakshminarayanan, Andrew M. Dai, Dustin Tran

  Access Paper or Ask Questions

The Break-Even Point on Optimization Trajectories of Deep Neural Networks

Feb 21, 2020
Stanislaw Jastrzebski, Maciej Szymczak, Stanislav Fort, Devansh Arpit, Jacek Tabor, Kyunghyun Cho, Krzysztof Geras

* Accepted as a spotlight at ICLR 2020. The last two authors contributed equally 

  Access Paper or Ask Questions

Deep Ensembles: A Loss Landscape Perspective

Dec 05, 2019
Stanislav Fort, Huiyi Hu, Balaji Lakshminarayanan

  Access Paper or Ask Questions

Emergent properties of the local geometry of neural loss landscapes

Oct 14, 2019
Stanislav Fort, Surya Ganguli

* 10 pages, 8 figures 

  Access Paper or Ask Questions

Large Scale Structure of Neural Network Loss Landscapes

Jun 11, 2019
Stanislav Fort, Stanislaw Jastrzebski

* Submitted for review 

  Access Paper or Ask Questions

Stiffness: A New Perspective on Generalization in Neural Networks

Jan 28, 2019
Stanislav Fort, Paweł Krzysztof Nowak, Srini Narayanan

  Access Paper or Ask Questions

Adaptive Quantum State Tomography with Neural Networks

Dec 17, 2018
Yihui Quek, Stanislav Fort, Hui Khoon Ng

* First two authors (Yihui Quek and Stanislav Fort) contributed equally. 13 pages, 10 figures 

  Access Paper or Ask Questions

The Goldilocks zone: Towards better understanding of neural network loss landscapes

Jul 06, 2018
Stanislav Fort, Adam Scherlis

* 14 pages, 14 figures. A subset of the paper accepted at Modern Trends in Nonconvex Optimization for Machine Learning workshop at the 35th International Conference on Machine Learning (ICML 2018) 

  Access Paper or Ask Questions

Towards understanding feedback from supermassive black holes using convolutional neural networks

Dec 02, 2017
Stanislav Fort

* 5 pages, 5 figures, accepted at Workshop on Deep Learning for Physical Sciences (DLPS 2017), NIPS 2017, Long Beach, CA, USA 

  Access Paper or Ask Questions

Gaussian Prototypical Networks for Few-Shot Learning on Omniglot

Aug 09, 2017
Stanislav Fort

  Access Paper or Ask Questions