### Current browse context:

stat.ML

### Change to browse by:

### References & Citations

# Statistics > Machine Learning

# Title: Practical Algorithms for Learning Near-Isometric Linear Embeddings

(Submitted on 1 Jan 2016 (v1), last revised 22 Apr 2016 (this version, v2))

Abstract: We propose two practical non-convex approaches for learning near-isometric, linear embeddings of finite sets of data points. Given a set of training points $\mathcal{X}$, we consider the secant set $S(\mathcal{X})$ that consists of all pairwise difference vectors of $\mathcal{X}$, normalized to lie on the unit sphere. The problem can be formulated as finding a symmetric and positive semi-definite matrix $\boldsymbol{\Psi}$ that preserves the norms of all the vectors in $S(\mathcal{X})$ up to a distortion parameter $\delta$. Motivated by non-negative matrix factorization, we reformulate our problem into a Frobenius norm minimization problem, which is solved by the Alternating Direction Method of Multipliers (ADMM) and develop an algorithm, FroMax. Another method solves for a projection matrix $\boldsymbol{\Psi}$ by minimizing the restricted isometry property (RIP) directly over the set of symmetric, postive semi-definite matrices. Applying ADMM and a Moreau decomposition on a proximal mapping, we develop another algorithm, NILE-Pro, for dimensionality reduction. FroMax is shown to converge faster for smaller $\delta$ while NILE-Pro converges faster for larger $\delta$. Both non-convex approaches are then empirically demonstrated to be more computationally efficient than prior convex approaches for a number of applications in machine learning and signal processing.

## Submission history

From: Hao-Jun Shi [view email]**[v1]**Fri, 1 Jan 2016 09:06:11 GMT (395kb,D)

**[v2]**Fri, 22 Apr 2016 21:47:56 GMT (488kb,D)

Link back to: arXiv, form interface, contact.