High Dimensional Linear Regression using Lattice Basis Reduction

Authors: Ilias Zadik, David Gamarnik

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

Reproducibility Variable Result LLM Response
Research Type Experimental In this section we present an experimental analysis of the ELO and LBR algorithms.
Researcher Affiliation Academia David Gamarnik Sloan School of Management Massachussetts Institute of Technology Cambridge, MA 02139 gamarnik@mit.edu; Ilias Zadik Operations Research Center Massachussetts Institute of Technology Cambridge, MA 02139 izadik@mit.edu
Pseudocode Yes Algorithm 1 Extended Lagarias-Odlyzko (ELO) Algorithm; Algorithm 2 Lattice Based Regression (LBR) Algorithm
Open Source Code No The paper does not provide any links to open-source code or explicitly state that the code for the described methodology is publicly available.
Open Datasets No Each entry of β is iid Unif ({1, 2, . . . , R = 100}). For 10 values of α (0, 3), specifically α {0.25, 0.5, 0.75, 1, 1.3, 1.6, 1.9, 2.25, 2.5, 2.75}, we generate the entries of X iid Unif {1, 2, 3, . . . , 2N} for N = p2 /2αn. We generate each entry of β w.p. 0.5 equal to zero and w.p. 0.5, Unif ({1, 2, . . . , R = 100}). We generate the entries of X iid U(0, 1) and of W iid U( σ, σ) for σ {0, e 20, e 12, e 4}.
Dataset Splits No The paper describes generating synthetic data for experiments and does not specify traditional train/validation/test dataset splits. It generates independent instances for evaluation.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU/GPU models, memory) used for running the experiments.
Software Dependencies No The paper does not mention specific software dependencies with version numbers required to replicate the experiments.
Experiment Setup Yes ELO algorithm: We focus on p = 30 features sample sizes n = 1, n = 10 and n = 30, R = 100 and zero-noise W = 0. Each entry of β is iid Unif ({1, 2, . . . , R = 100}). For 10 values of α (0, 3), specifically α {0.25, 0.5, 0.75, 1, 1.3, 1.6, 1.9, 2.25, 2.5, 2.75}, we generate the entries of X iid Unif {1, 2, 3, . . . , 2N} for N = p2 /2αn. For each combination of n, α we generate 20 independent instances of inputs. LBR algorithm: We focus on p = 30 features, n = 10 samples, Q = 1 and R = 100. We generate each entry of β w.p. 0.5 equal to zero and w.p. 0.5, Unif ({1, 2, . . . , R = 100}). We generate the entries of X iid U(0, 1) and of W iid U( σ, σ) for σ {0, e 20, e 12, e 4}. We generate 20 independent instances for any combination of σ and truncation level N.