Online Risk-Averse Submodular Maximization

Authors: Tasuku Soma, Yuichi Yoshida

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

Reproducibility Variable Result LLM Response
Research Type Experimental 5 Experiments In this section, we show our experimental results. ... Datasets. We consider two sensing models and generated three datasets. ... Results. Figure 1 shows how the CVa R changes as T increases. ... Figure 2 shows how the CVa R changes as the budget B increases.
Researcher Affiliation Academia Tasuku Soma1,2 , Yuichi Yoshida3 1 The University of Tokyo 2 Massachusetts Institute of Technology 3 National Institute of Informatics
Pseudocode Yes Algorithm 1 STOCHASTICRASCAL ... Algorithm 2 SMOOTHGRAD(x, τ, u, Z) ... Algorithm 3 SMOOTHTAU(x, u, Z) ... Algorithm 4 Online algorithm for maximizing a monotone submodular set function subject to a matroid constraint.
Open Source Code No The paper does not provide any statements or links indicating that the source code for the described methodology is publicly available.
Open Datasets Yes Datasets. We consider two sensing models and generated three datasets. ... Net Science, a collaboration network of 1,461 network scientists, and Euro Road, a network of 1,174 European cities and the roads between them. For both networks, we set λ = 5 and p = 0.01, and we generated 1,000 scenarios. ... The second model, known as the Battle of Water Sensor Networks (BWSN) [Ostfeld and others, 2008]...
Dataset Splits No The paper states it generated 1,000 scenarios and ran methods on them, and ran its online method on 20,000 samples drawn from these scenarios, but it does not specify explicit training, validation, or test dataset splits (e.g., percentages or counts) or reference standard pre-defined splits.
Hardware Specification Yes The experiments were conducted on a Linux server with Intel Xeon Gold 6242 (2.8GHz) and 384GB of main memory.
Software Dependencies No The paper does not specify any software dependencies with their version numbers.
Experiment Setup Yes In all the experiments, the parameter α of CVa R was set to 0.1. ... For both networks, we set λ = 5 and p = 0.01... We set p = 0.001 and generated 1,000 scenarios. ... Ours (batch size = 32) Ours (batch size = 64) Ours (batch size = 128) Ours (batch size = 256)