= Ali Rahimi
{c}
<Ali Rahimi> coauthored <Random Features for Large-Scale Kernel Machines by Ali Rahimi and Benjamin Recht> in 2007. With <Benjamin Recht>, he proposed explicit randomized maps whose <inner product>[inner products] approximate a chosen <positive-semidefinite kernel>[kernel], enabling familiar fast linear methods in the transformed feature space.<Random Features for Large-Scale Kernel Machines by Ali Rahimi and Benjamin Recht>[abstract and §3]{ref}
The <Random Fourier feature> mechanism samples <frequency>[frequencies] from a shift-invariant <positive-semidefinite kernel>[kernel]'s spectral representation. This separates the cost of constructing a <feature map> from the cost of repeatedly evaluating every pairwise <positive-semidefinite kernel>[kernel] value. Accuracy and computation depend on the feature count; randomness does not eliminate <approximation error> or the need to validate the learned predictor.
Back to article page
