Online MAP Inference of Determinantal Point Processes

Authors: Aditya Bhaskara, Amin Karbasi, Silvio Lattanzi, Morteza Zadimoghaddam

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments demonstrating the efficiency of online methods for MAP inference. We demonstrate how even the simple algorithm ONLINE-LS finds solutions that compete favorably with offline algorithms (that store the entire dataset in memory).
Researcher Affiliation Collaboration Aditya Bhaskara School of Computing University of Utah bhaskaraaditya@gmail.com Amin Karbasi School of Engineering & Applied Science Yale University amin.karbasi@yale.edu Silvio Lattanzi Google Research Zürich silviol@google.com Morteza Zadimoghaddam Google Research Cambridge zadim@google.com
Pseudocode Yes Algorithm 1 Local Search with Stash (ONLINE-LS) and Algorithm 2 Online Coreset for Additive Error Approximation (ONLINE-DPP) are presented.
Open Source Code No The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes In our experiments we consider three standard datasets: the Spambase dataset [Dua and Graff, 2017], the Statlog(or Shuttle) dataset [Dua and Graff, 2017] and the Pen-Based Recognition dataset [Dua and Graff, 2017].
Dataset Splits No The paper mentions using standard datasets but does not specify the train/validation/test splits, nor does it refer to predefined splits with citations.
Hardware Specification No All our experiments have been carried out on a standard desktop computer.
Software Dependencies No The paper does not provide specific software dependencies or version numbers.
Experiment Setup Yes In Figure 1 we show a comparison of the three algorithms on the Spambase dataset and the Statlog dataset for k = 8 and = 0.1. ... we report how the number of swaps and quality of the solution change as changes (experiments on other datasets are available in supplementary material).