Mixed-Features Vectors and Subspace Splitting

Authors: Alejandro Pimentel-Alarcón, Daniel L. Pimentel-Alarcón

ICLR 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our extensive experiments outline the performance of our algorithms, and in lack of other SS algorithms, for reference we compare against methods for tightly related problems, like robust matched subspace detection and maximum feasible subsystem, which are special simpler cases of SS.
Researcher Affiliation Academia Alejandro Pimentel-Alarcón IIMAS, Universidad Nacional Autónoma de México, UNAM Mexico City, Mexico pimentel@comunidad.unam.mx Daniel Pimentel-Alarcón Department of Biostatistics Wisconsin Institute for Discovery UW-Madison, WI, 53715 pimentelalar@wisc.edu
Pseudocode No The paper describes the algorithms in prose but does not provide structured pseudocode blocks or figures labeled 'Algorithm'.
Open Source Code Yes In the interest of reproducibility, all our code is available here [100].
Open Datasets No In our experiments we first generate matrices U1, . . . , UK Rd r with i.i.d. N(0, 1) entries, to use as bases of U1, . . . , UK, with d = (K + 1)|Ωk|. Similarly, we generate θ1, . . . , θK Rr, also with i.i.d. N(0, 1) entries, to use as coefficient vectors.
Dataset Splits No The paper describes the synthetic data generation process and how parameters like K, r, |Ωk|, p, σ are varied for experiments, but it does not specify fixed training, validation, or test dataset splits in terms of percentages or sample counts.
Hardware Specification No The paper does not provide specific details about the hardware used for running the experiments (e.g., CPU/GPU models, memory).
Software Dependencies No The paper does not list specific software dependencies with version numbers (e.g., Python 3.x, PyTorch 1.x).
Experiment Setup No The paper describes how the synthetic data is generated and the ranges for parameters like K, r, |Ωk|, p, σ. However, for the K-SPLITS algorithm, it mentions 'τ(σ) is a tuning thresholding parameter that depends on the noise level' without providing a specific value or formula for τ(σ), which is critical for reproducibility.