Sketching Method for Large Scale Combinatorial Inference

Authors: Wei Sun, Junwei Lu, Han Liu

NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We validate our theory and method through both synthetic simulations and a real application in neuroscience.
Researcher Affiliation Academia Will Wei Sun Department of Management Science University of Miami wsun@bus.miami.eduJunwei Lu Department of Biostatistics Harvard University junweilu@hsph.harvard.eduHan Liu Department of Computer Science Northwestern University hanliu@northwestern.edu
Pseudocode Yes Algorithm 1 Fast Connectivity Test
Open Source Code No The paper does not provide an explicit statement about releasing source code or a link to a code repository for the methodology.
Open Datasets Yes We apply our sketching-based inferential methods to an Neuroimaging study conducted by [28].
Dataset Splits Yes Throughout all our experiments, we tune λ in bθ1 via cross-validation and use the same λ for the rest bθj.
Hardware Specification No The paper discusses computational time but does not specify the hardware (e.g., CPU, GPU models) used for experiments.
Software Dependencies No The paper mentions algorithms and methods like node-wise regression, CLIME, graphical lasso, and Bron-Kerbosch, but does not specify any software names with version numbers for implementation.
Experiment Setup Yes Throughout all our experiments, we tune λ in bθ1 via cross-validation and use the same λ for the rest bθj. We then estimate s as bs = bθ1 0 and estimate ϵ as 2/(bs d). We use the theoretical rate for τ and set τ = 0.5 p log(d)/n.