Corpus ID: 218972066

Demystifying Orthogonal Monte Carlo and Beyond

@article{Lin2020DemystifyingOM,
  title={Demystifying Orthogonal Monte Carlo and Beyond},
  author={H. Lin and Haoxian Chen and Tianyi Zhang and Cl{\'e}ment Laroche and Krzysztof Choromanski},
  journal={ArXiv},
  year={2020},
  volume={abs/2005.13590}
}
Orthogonal Monte Carlo (OMC) is a very effective sampling algorithm imposing structural geometric conditions (orthogonality) on samples for variance reduction. Due to its simplicity and superior performance as compared to its Quasi Monte Carlo counterparts, OMC is used in a wide spectrum of challenging machine learning applications ranging from scalable kernel methods to predictive recurrent neural networks, generative models and reinforcement learning. However theoretical understanding of the… Expand
3 Citations
Rethinking Attention with Performers
  • 58
  • PDF
Masked Language Modeling for Proteins via Linearly Scalable Long-Context Transformers
  • 10
  • PDF

References

SHOWING 1-10 OF 45 REFERENCES
Structured Monte Carlo Sampling for Nonisotropic Distributions via Determinantal Point Processes
  • 3
  • PDF
Unifying Orthogonal Monte Carlo Methods
  • 17
  • PDF
Geometrically Coupled Monte Carlo Sampling
  • 21
  • PDF
Quasi-Monte Carlo Feature Maps for Shift-Invariant Kernels
  • 116
  • Highly Influential
  • PDF
KAMA-NNs: Low-dimensional Rotation Based Neural Networks
  • 7
  • PDF
Orthogonal Random Features
  • 120
  • PDF
Monte Carlo with Determinantal Point Processes
  • 51
  • PDF
The Geometry of Random Features
  • 22
  • PDF
On two ways to use determinantal point processes for Monte Carlo integration
  • 11
  • PDF
Recycling Randomness with Structure for Sublinear time Kernel Expansions
  • 34
  • PDF
...
1
2
3
4
5
...