The Geometry of Random Features

By Krzysztof Marcin Choromanski, Mark Rowland, Tamás Sarlós, Vikas Sindhwani, Richard E. Turner, Adrian Weller. 2018

43 citations recorded by the source corpus.

Subfields: Representation & Self-Supervised Learning, Scaling & Learning Theory

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Representation & Self-Supervised Learning

1. Krzysztof Marcin Choromanski
2. Adrian Weller
3. Tamás Sarlós
4. Mark Rowland
5. Wenyu Chen
6. Valerii Likhosherstov
7. Haoxian Chen
8. David Dohan

In Scaling & Learning Theory

1. Adrian Weller
2. Xingyou Song

Citations in the corpus

• Rethinking Attention with Performers (2020, background): “It turns out that: The proof of that fact can be found in the supplement of (Choromanski et al., 2018b), yet we provide it below for completeness and the convenience of the Reader: Lemma 3.”
• Masked Language Modeling for Proteins via Linearly Scalable Long-Context Transformers (2020, methodology): “2) and a radial basis function (RBF) kernel [12] K with corresponding spectral distribution Ω (e.”
• Unifying Orthogonal Monte Carlo Methods (2019, methodology): “…reduction (Choromanski et al., 2017), locality-sensitive hashing (Andoni et al., 2015), random feature approximations to kernel methods such as Gaussian processes (Choromanski et al., 2018a) and support vector machines (Yu et al., 2016), and black-box optimization (Choromanski et al., 2018b).”
• Demystifying Orthogonal Monte Carlo and Beyond (2020, background): “General RBF kernels: Even more interestingly, Theorem 2 enables us to give first strong concentration results for all RBF kernels, avoiding very cumbersome technical requirements regarding tails of the corresponding spectral distributions (see: [15, 11]).”