research
∙
02/05/2023
KDEformer: Accelerating Transformers via Kernel Density Estimation
Dot-product attention mechanism plays a crucial role in modern deep arch...
share
research
∙
09/09/2022
Fast Neural Kernel Embeddings for General Activations
Infinite width limit has shed light on generalization and optimization a...
research
∙
02/25/2022
Near Optimal Reconstruction of Spherical Harmonic Expansions
We propose an algorithm for robust recovery of the spherical harmonic ex...
research
∙
02/09/2022
Leverage Score Sampling for Tensor Product Matrices in Input Sparsity Time
We give an input sparsity time sampling algorithm for spectrally approxi...
research
∙
02/07/2022
Random Gegenbauer Features for Scalable Kernel Methods
We propose efficient random features for approximating a new and rich cl...
research
∙
07/15/2021
Sparse Fourier Transform by traversing Cooley-Tukey FFT computation graphs
Computing the dominant Fourier coefficients of a vector is a common task...
research
∙
06/15/2021
Scaling Neural Tangent Kernels via Sketching and Random Features
The Neural Tangent Kernel (NTK) characterizes the behavior of infinitely...
research
∙
04/01/2021
Learning with Neural Tangent Kernels in Near Input Sparsity Time
The Neural Tangent Kernel (NTK) characterizes the behavior of infinitely...
research
∙
07/08/2020
Near Input Sparsity Time Kernel Embeddings via Adaptive Sampling
To accelerate kernel methods, we propose a near input sparsity time algo...
research
∙
03/21/2020
Scaling up Kernel Ridge Regression via Locality Sensitive Hashing
Random binning features, introduced in the seminal paper of Rahimi and R...
research
∙
09/03/2019
Oblivious Sketching of High-Degree Polynomial Kernels
Kernel methods are fundamental tools in machine learning that allow dete...
research
∙
02/27/2019
Dimension-independent Sparse Fourier Transform
The Discrete Fourier Transform (DFT) is a fundamental computational prim...
research
∙
12/20/2018
A Universal Sampling Method for Reconstructing Signals with Simple Fourier Transforms
Reconstructing continuous signals from a small number of discrete sample...
research
∙
08/06/2018
Beyond 1/2-Approximation for Submodular Maximization on Massive Data Streams
Many tasks in machine learning and data mining, such as data diversifica...
research
∙
04/26/2018