arXiv subjects

Daniel Mckenzie

Publications and source records attributed to Daniel Mckenzie.

7 recordsLinked to original sources

arXiv2020

Who killed Lilly Kane? A case study in applying knowledge graphs to crime fiction

We present a preliminary study of a knowledge graph created from season one of the television show Veronica Mars, which follows the eponymous young private investigator as she attempts to solve the murder of her best friend Lilly Kane. We discuss various techniques for mining the knowledge graph for clues and potential suspects. We also discuss best practice for collaboratively constructing knowledge graphs from television shows.

cs.LG↗

arXiv2020

A One-bit, Comparison-Based Gradient Estimator

We study zeroth-order optimization for convex functions where we further assume that function evaluations are unavailable. Instead, one only has access to a $\textit{comparison oracle}$, which given two points $x$ and $y$ returns a single bit of information indicating which point has larger function value, $f(x)$ or $f(y)$. By treating the gradient as an unknown signal to be recovered, we show how one can use tools from one-bit compressed sensing to construct a robust and reliable estimator of the normalized gradient. We then propose an algorithm, coined SCOBO, that uses this estimator within a gradient descent scheme. We show that when $f(x)$ has some low dimensional structure that can be exploited, SCOBO outperforms the state-of-the-art in terms of query complexity. Our theoretical claims are verified by extensive numerical experiments.

math.OC↗

arXiv2020

Zeroth-Order Regularized Optimization (ZORO): Approximately Sparse Gradients and Adaptive Sampling

We consider the problem of minimizing a high-dimensional objective function, which may include a regularization term, using (possibly noisy) evaluations of the function. Such optimization is also called derivative-free, zeroth-order, or black-box optimization. We propose a new $\textbf{Z}$eroth-$\textbf{O}$rder $\textbf{R}$egularized $\textbf{O}$ptimization method, dubbed ZORO. When the underlying gradient is approximately sparse at an iterate, ZORO needs very few objective function evaluations to obtain a new iterate that decreases the objective function. We achieve this with an adaptive, randomized gradient estimator, followed by an inexact proximal-gradient scheme. Under a novel approximately sparse gradient assumption and various different convex settings, we show the (theoretical and empirical) convergence rate of ZORO is only logarithmically dependent on the problem dimension. Numerical experiments show that ZORO outperforms the existing methods with similar assumptions, on both synthetic and real datasets.

math.OC↗

arXiv2019

Power Weighted Shortest Paths for Clustering Euclidean Data

We study the use of power weighted shortest path distance functions for clustering high dimensional Euclidean data, under the assumption that the data is drawn from a collection of disjoint low dimensional manifolds. We argue, theoretically and experimentally, that this leads to higher clustering accuracy. We also present a fast algorithm for computing these distances.

cs.LG↗

arXiv2018

On the neighborliness of dual flow polytopes of quivers

In this note we investigate under which conditions the dual of the flow polytope (henceforth referred to as the `dual flow polytope') of a quiver is k-neighborly, for generic weights near the canonical weight. We provide a lower bound on k, depending only on the edge connectivity of the underlying graph, for such weights. In the case where the canonical weight is in fact generic, we explicitly determine a vertex presentation of the dual flow polytope. Finally, we specialize our results to the case of complete, bipartite quivers where we show that the canonical weight is indeed generic, and are able to provide an improved bound on k. Hence, we are able to produce many new examples of high-dimensional, k-neighborly polytopes.

math.CO↗

arXiv2018

Compressive Sensing for cut improvement and local clustering

We show how one can phrase the cut improvement problem for graphs as a sparse recovery problem, whence one can use algorithms originally developed for use in compressive sensing (such as SubspacePursuit or CoSaMP) to solve it. We show that this approach to cut improvement is fast, both in theory and practice and moreover enjoys statistical guarantees of success when applied to graphs drawn from probabilistic models such as the Stochastic Block Model. Using this new cut improvement approach, which we call ClusterPursuit, as an algorithmic primitive we then propose new methods for local clustering and semi-supervised clustering, which enjoy similar guarantees of success and speed. Finally, we verify the promise of our approach with extensive numerical benchmarking.

cs.IT↗

arXiv2017

A Compressive Sensing Approach to Community Detection with Applications

The community detection problem for graphs asks one to partition the n vertices V of a graph G into k communities, or clusters, such that there are many intracluster edges and few intercluster edges. Of course this is equivalent to finding a permutation matrix P such that, if A denotes the adjacency matrix of G, then PAP^T is approximately block diagonal. As there are k^n possible partitions of n vertices into k subsets, directly determining the optimal clustering is clearly infeasible. Instead one seeks to solve a more tractable approximation to the clustering problem. In this paper we reformulate the community detection problem via sparse solution of a linear system associated with the Laplacian of a graph G and then develop a two-stage approach based on a thresholding technique and a compressive sensing algorithm to find a sparse solution which corresponds to the community containing a vertex of interest in G. Crucially, our approach results in an algorithm which is able to find a single cluster of size n_0 in O(nlog(n)n_0) operations and all k clusters in fewer than O(n^2ln(n)) operations. This is a marked improvement over the classic spectral clustering algorithm, which is unable to find a single cluster at a time and takes approximately O(n^3) operations to find all k clusters. Moreover, we are able to provide robust guarantees of success for the case where G is drawn at random from the Stochastic Block Model, a popular model for graphs with clusters. Extensive numerical results are also provided, showing the efficacy of our algorithm on both synthetic and real-world data sets.

cs.IT↗