Sparse projections onto the simplex
arXiv:1206.1529
Abstract
Most learning methods with rank or sparsity constraints use convex relaxations, which lead to optimization with the nuclear norm or the -norm. However, several important learning applications cannot benefit from this approach as they feature these convex norms as constraints in addition to the non-convex rank and sparsity constraints. In this setting, we derive efficient sparse projections onto the simplex and its extension, and illustrate how to use them to solve high-dimensional learning problems in quantum tomography, sparse density estimation and portfolio selection with non-convex constraints.
9 Pages
References in corpus (4)
Cited by in corpus (19)
- Hyperspectral unmixing with spectral variability using adaptive bundles and double sparsity
- An Empirical-Bayes Approach to Recovering Linearly Constrained Non-Negative Sparse Signals
- Efficient Solvers for Sparse Subspace Clustering
- On Geometric Analysis of Affine Sparse Subspace Clustering
- Joint Inference of Multiple Label Types in Large Networks
- Phenotyping using Structured Collective Matrix Factorization of Multi--source EHR Data
- Efficient Marginalization of Discrete and Structured Latent Variables via Sparsity
- Provable quantum state tomography via non-convex methods
- Learning with Fenchel-Young Losses
- DARC: Differentiable ARchitecture Compression
- Fast Spectral Unmixing based on Dykstra's Alternating Projection
- Positively Weighted Kernel Quadrature via Subsampling
- Peaceman-Rachford splitting for a class of nonconvex optimization problems
- Sparse regular variation
- Wasserstein k-means with sparse simplex projection
- A Lagrange-Newton Algorithm for Sparse Nonlinear Programming
- Double spike Dirichlet priors for structured weighting
- A difference-of-convex approach for split feasibility with applications to matrix factorizations and outlier detection
- Estimating Sparse Discrete Distributions Under Local Privacy and Communication Constraints