Learning sparse features can lead to overfitting in neural networks
From MaRDI portal
Recommendations
- Redundant representations help generalization in wide neural networks
- More is Less: Inducing Sparsity via Overparameterization
- Deep learning: a statistical viewpoint
- Scaling description of generalization with number of parameters in deep learning
- Sparse deep neural networks using \(L_{1,\infty}\)-weight normalization
Cites work
- A mean field view of the landscape of two-layer neural networks
- Asymptotic learning curves of kernel methods: empirical data versus teacher–student paradigm
- Atomic Decomposition by Basis Pursuit
- Breaking the curse of dimensionality with convex neural networks
- Disentangling feature and lazy training in deep neural networks
- Distance-based classification with Lipschitz functions
- Generalization error rates in kernel regression: the crossover from the noiseless to noisy regime*
- Geometric compression of invariant manifolds in neural networks
- Group invariance, stability to deformations, and complexity of deep convolutional representations
- Landscape and training regimes in deep learning
- Mean field analysis of neural networks: a law of large numbers
- On representer theorems and convex regularization
- Prevalence of neural collapse during the terminal phase of deep learning training
- Scaling description of generalization with number of parameters in deep learning
- Sparse optimization on measures with over-parameterized gradient descent
- Spherical harmonics and approximations on the unit sphere. An introduction
- Spherical harmonics in \(p\) dimensions
- When do neural networks outperform kernel methods?*
Cited in
(4)
This page was built for publication: Learning sparse features can lead to overfitting in neural networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6611436)
