How to Solve Fair k-Center in Massive Data Models

Authors: Ashish Chiplunkar, Sagar Kale, Sivaramakrishnan Natarajan Ramamoorthy

ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We perform experiments on real and synthetic datasets and show that our algorithms are as fast as the linear-time algorithm of Kleindessner et al., while achieving improved approximation ratio, which matches that of Chen et al.
Researcher Affiliation Collaboration 1Indian Institute of Technology Delhi 2University of Vienna, Faculty of Computer Science 3Theorem LP.
Pseudocode Yes Algorithm 1 Two-pass algorithm
Open Source Code Yes We make our code available on Git Hub2. 2https://github.com/sagark4/fair_k_center
Open Datasets Yes We use three real world datasets: Celeb-A (Liu et al., 2015), Sushi (sus), and Adult (Kohavi & Becker), with n = 1000 by selecting the first 1000 data points (see Table 1).
Dataset Splits No The paper mentions using specific datasets but does not provide details on how they were split into training, validation, or test sets. For example, it states 'We use three real world datasets: Celeb-A (Liu et al., 2015), Sushi (sus), and Adult (Kohavi & Becker), with n = 1000 by selecting the first 1000 data points' without specifying data partitioning for reproduction.
Hardware Specification Yes All experiments are run on HP Elite Book 840 G6 with Intel R Core TM i7-8565U CPU 1.80GHz having 4 cores and 15.5 Gi B of RAM, running Ubuntu 18.04 and Anaconda.
Software Dependencies No The paper mentions several software components like 'Ubuntu 18.04 and Anaconda', 'Python s multiprocessing library', 'Keras', 'VGG16', and 'Python package NetworkX', but it does not specify version numbers for these, which is required for reproducibility.
Experiment Setup Yes For all algorithms (except Kleindessner et al. s), we use ε = 0.1. For the distributed algorithm, we use block size of 25, i.e., the number of processors are 1000/25 = 40: theoretically, using n processor gives maximum speedup.