Corpus ID: 107263

Random Features for Compositional Kernels

@article{Daniely2017RandomFF,
  title={Random Features for Compositional Kernels},
  author={Amit Daniely and Roy Frostig and Vineet Gupta and Y. Singer},
  journal={ArXiv},
  year={2017},
  volume={abs/1703.07872}
}
We describe and analyze a simple random feature scheme (RFS) from prescribed compositional kernels. The compositional kernels we use are inspired by the structure of convolutional neural networks and kernels. The resulting scheme yields sparse and efficiently computable features. Each random feature can be represented as an algebraic expression over a small number of (random) paths in a composition tree. Thus, compositional random features can be stored compactly. The discrete nature of the… Expand
11 Citations

Paper Mentions

Kernel-based Translations of Convolutional Networks
  • 4
  • PDF
Group Invariance, Stability to Deformations, and Complexity of Deep Convolutional Representations
  • 38
  • PDF
Data-dependent compression of random features for large-scale kernel approximation
  • 12
  • PDF
Group Invariance and Stability to Deformations of Deep Convolutional Representations
  • 8
  • PDF
SGD Learns the Conjugate Kernel Class of the Network
  • 121
  • PDF
Porcupine Neural Networks: Approximating Neural Network Landscapes
  • 5
  • PDF
End-to-end Learning, with or without Labels
  • PDF
...
1
2
...

References

SHOWING 1-10 OF 22 REFERENCES
Spherical Random Features for Polynomial Kernels
  • 49
  • PDF
On Random Weights and Unsupervised Feature Learning
  • 351
  • PDF
Random Feature Maps for Dot Product Kernels
  • 184
  • PDF
Random Features for Large-Scale Kernel Machines
  • 2,298
  • Highly Influential
  • PDF
Kernel Methods for Deep Learning
  • 468
  • Highly Influential
  • PDF
Convolutional Kernel Networks
  • 287
  • Highly Influential
  • PDF
End-to-End Kernel Learning with Supervised Convolutional Kernel Networks
  • J. Mairal
  • Computer Science, Mathematics
  • NIPS
  • 2016
  • 81
  • Highly Influential
  • PDF
Object recognition with hierarchical kernel descriptors
  • 247
  • PDF
The pyramid match kernel: discriminative classification with sets of image features
  • K. Grauman, Trevor Darrell
  • Mathematics, Computer Science
  • Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1
  • 2005
  • 1,574
  • PDF
Toward Deeper Understanding of Neural Networks: The Power of Initialization and a Dual View on Expressivity
  • 194
  • PDF
...
1
2
3
...