Differentiable Submodular Maximization

Authors: Sebastian Tschiatschek, Aytunc Sahin, Andreas Krause

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

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate the effectiveness of our approach for jointly learning and optimizing on synthetic maximum cut data, and on real world applications such as product recommendation and image collection summarization.
Researcher Affiliation Collaboration Sebastian Tschiatschek1, Aytunc Sahin2 and Andreas Krause2 1 Microsoft Research Cambridge 2 ETH Zurich
Pseudocode Yes Algorithm 1 PD2GREEDY: Probabilistic diff. double-greedy
Open Source Code No The paper does not provide an explicit statement or link for the release of its source code.
Open Datasets Yes We consider the Amazon baby registry data [Gillenwater et al., 2014].
Dataset Splits Yes The data is processed using 10-fold cross-validation.
Hardware Specification No The paper does not provide specific details about the hardware used for running experiments.
Software Dependencies No The paper mentions software like Adam optimizer and Gurobi but does not provide specific version numbers for any software dependencies.
Experiment Setup Yes For optimization, we used Adam [Kingma and Ba, 2015] with batch size of 16 and initial learning rate 0.02.