Stable rank normalization for improved generalization in neural networks and GANs

Stable rank normalization for improved generalization in neural networks and GANs

Exciting new work on generalization bounds for neural networks (NN) given by Bartlett et al. (2017); Neyshabur et al. (2018) closely depend on two parameter- dependant quantities: the Lipschitz constant upper bound and the stable rank (a softer version of rank). Even though these bounds typically ha...

Cur síos iomlán

Sonraí bibleagrafaíochta
Príomhchruthaitheoirí:	Sanyal, A, Torr, P, Dokania, P
Formáid:	Conference item
Teanga:	English
Foilsithe / Cruthaithe:	International Conference on Learning Representations 2020

Míreanna comhchosúla

On using focal loss for neural network calibration
de réir: Mukhoti, J, et al.
Foilsithe / Cruthaithe: (2020)

Calibrating deep neural networks using focal loss
de réir: Mukhoti, J, et al.
Foilsithe / Cruthaithe: (2020)

On Generalization Bounds for Neural Networks with Low Rank Layers
de réir: Pinto, Andrea, et al.
Foilsithe / Cruthaithe: (2024)

Proximal mean-field for neural network quantization
de réir: Ajanthan, T, et al.
Foilsithe / Cruthaithe: (2020)

Are vision transformers always more robust than convolutional neural networks?
de réir: Pinto, F, et al.
Foilsithe / Cruthaithe: (2021)

Mix-MaxEnt: improving accuracy and uncertainty estimates of deterministic neural networks
de réir: Pinto, F, et al.
Foilsithe / Cruthaithe: (2021)

Continual learning in low-rank orthogonal subspaces
de réir: Chaudhry, A, et al.
Foilsithe / Cruthaithe: (2020)

How benign is benign overfitting?
de réir: Sanyal, A, et al.
Foilsithe / Cruthaithe: (2021)

Mirror Descent view for Neural Network quantization
de réir: Ajanthan, T, et al.
Foilsithe / Cruthaithe: (2021)

Progressive skeletonization: trimming more fat from a network at initialization
de réir: de Jorge, P, et al.
Foilsithe / Cruthaithe: (2020)

Lyapunov-stable neural-network control
de réir: Dai, Hongkai, et al.
Foilsithe / Cruthaithe: (2022)

Improving generalization of neural network using length as discriminant
de réir: Siraj, Fadzilah, et al.
Foilsithe / Cruthaithe: (1999)

Normal Modified Stable Processes.
de réir: Shephard, N, et al.
Foilsithe / Cruthaithe: (2001)

Normal modified stable processes
de réir: Shephard, N, et al.
Foilsithe / Cruthaithe: (2001)

Neural SDEs as infinite-dimensional GANs
de réir: Kidger, P, et al.
Foilsithe / Cruthaithe: (2021)

Rough Neural Networks Architecture For Improving Generalization In Pattern Recognition
de réir: Ali Adlan, Hanan Hassan
Foilsithe / Cruthaithe: (2004)

Improving generalization of neural networks using multilayer perceptron discriminants
de réir: Siraj, Fadzilah, et al.
Foilsithe / Cruthaithe: (2002)

Speeding up convolutional neural networks with low rank expansions
de réir: Jaderberg, M, et al.
Foilsithe / Cruthaithe: (2014)

Speeding up convolutional neural networks with low rank expansions
de réir: Jaderberg, M, et al.
Foilsithe / Cruthaithe: (2014)

On the robustness of quality measures for GANs
de réir: Alfarra, M, et al.
Foilsithe / Cruthaithe: (2022)

Realistic face generation using deep neural networks (StyleGAN)
de réir: Toh, Wilson Chin Shen
Foilsithe / Cruthaithe: (2021)

On the existence of generalized rank weights
de réir: Oggier, Frederique, et al.
Foilsithe / Cruthaithe: (2013)

Generalized Low-Rank Approximations
de réir: Srebro, Nathan, et al.
Foilsithe / Cruthaithe: (2004)

What makes and breaks safety fine-tuning? a mechanistic study
de réir: Jain, S, et al.
Foilsithe / Cruthaithe: (2024)

Stable Rank-Adaptive Dynamically Orthogonal Runge–Kutta Schemes
de réir: Charous, Aaron, et al.
Foilsithe / Cruthaithe: (2024)

Varsities improve in rank
de réir: Menon, Sandhya
Foilsithe / Cruthaithe: (2021)

Varsities Improve On Rankings
de réir: The Star, Choi Tuck Wo
Foilsithe / Cruthaithe: (2008)

Stable homotopy and generalized homology/
de réir: 447926 Adams, John Frank
Foilsithe / Cruthaithe: (1974)

SGD Noise and Implicit Low-Rank Bias in Deep Neural Networks
de réir: Galanti, Tomer, et al.
Foilsithe / Cruthaithe: (2022)

Certifying ensembles: a general certification theory with s-lipschitzness
de réir: Petrov, A, et al.
Foilsithe / Cruthaithe: (2023)

Interpolating convolutional neural networks using batch normalization
de réir: Data, G, et al.
Foilsithe / Cruthaithe: (2018)

Multi-agent diverse generative adversarial networks
de réir: Ghosh, A, et al.
Foilsithe / Cruthaithe: (2018)

Stable and efficient computation of generalized polar decompositions
de réir: Benner, P, et al.
Foilsithe / Cruthaithe: (2022)

On Neural Network Pruning’s Effect on Generalization
de réir: Jin, Tian
Foilsithe / Cruthaithe: (2023)

On the Expressiveness and Generalization of Hypergraph Neural Networks
de réir: Luo, Zhezheng
Foilsithe / Cruthaithe: (2023)

Learning and generalization : with applications to neural networks /
de réir: 378518 Vidyasagar, M. Mathukumalli), 1947-
Foilsithe / Cruthaithe: (2003)

Development of low normal force contact with stable performance
de réir: Bhuiyan, Moinul
Foilsithe / Cruthaithe: (2010)

GeoNet++: Iterative geometric neural network with edge-aware refinement for joint depth and surface normal estimation
de réir: Qi, X, et al.
Foilsithe / Cruthaithe: (2020)

Interpreting neural network judgments via minimal, stable, and symbolic corrections
de réir: Solar Lezama, Armando, et al.
Foilsithe / Cruthaithe: (2021)

Improving generalization of neural networks using MLP discriminant based on multiple classifiers failures
de réir: Siraj, Fadzilah, et al.
Foilsithe / Cruthaithe: (2010)