Matches in SemOpenAlex for { <https://semopenalex.org/work/W3187304896> ?p ?o ?g. }
- W3187304896 abstract "As the field of machine learning for combinatorial optimization advances, traditional problems are resurfaced and readdressed through this new perspective. The overwhelming majority of the literature focuses on small graph problems, while several real-world problems are devoted to large graphs. Here, we focus on two such problems that are related: influence estimation, a #P-hard counting problem, and influence maximization, an NP-hard problem. We develop GLIE, a Graph Neural Network (GNN) that inherently parameterizes an upper bound of influence estimation and train it on small simulated graphs. Experiments show that GLIE can provide accurate predictions faster than the alternatives for graphs 10 times larger than the train set. More importantly, it can be used on arbitrary large graphs for influence maximization, as the predictions can rank effectively seed sets even when the accuracy deteriorates. To showcase this, we propose a version of a standard Influence Maximization (IM) algorithm where we substitute traditional influence estimation with the predictions of GLIE.We also transfer GLIE into a reinforcement learning model that learns how to choose seeds to maximize influence sequentially using GLIE's hidden representations and predictions. The final results show that the proposed methods surpasses a previous GNN-RL approach and perform on par with a state-of-the-art IM algorithm." @default.
- W3187304896 created "2021-08-16" @default.
- W3187304896 creator A5057695979 @default.
- W3187304896 creator A5079375197 @default.
- W3187304896 creator A5083139588 @default.
- W3187304896 creator A5087251228 @default.
- W3187304896 date "2021-08-10" @default.
- W3187304896 modified "2023-09-27" @default.
- W3187304896 title "Learning to Maximize Influence" @default.
- W3187304896 cites W1650780448 @default.
- W3187304896 cites W1931400479 @default.
- W3187304896 cites W1983275606 @default.
- W3187304896 cites W1988717476 @default.
- W3187304896 cites W2008620264 @default.
- W3187304896 cites W2035165116 @default.
- W3187304896 cites W2042123098 @default.
- W3187304896 cites W2061820396 @default.
- W3187304896 cites W2108858998 @default.
- W3187304896 cites W2121863487 @default.
- W3187304896 cites W2132801025 @default.
- W3187304896 cites W2139297408 @default.
- W3187304896 cites W2141403143 @default.
- W3187304896 cites W2144629587 @default.
- W3187304896 cites W2145339207 @default.
- W3187304896 cites W2155968351 @default.
- W3187304896 cites W2487770199 @default.
- W3187304896 cites W2507756961 @default.
- W3187304896 cites W2607264901 @default.
- W3187304896 cites W2621117388 @default.
- W3187304896 cites W2808189760 @default.
- W3187304896 cites W2912555327 @default.
- W3187304896 cites W2952332632 @default.
- W3187304896 cites W2962773920 @default.
- W3187304896 cites W2962795549 @default.
- W3187304896 cites W2963122331 @default.
- W3187304896 cites W2963382544 @default.
- W3187304896 cites W2963867952 @default.
- W3187304896 cites W2980930344 @default.
- W3187304896 cites W2997102038 @default.
- W3187304896 cites W3006978394 @default.
- W3187304896 cites W3028110392 @default.
- W3187304896 cites W3035768633 @default.
- W3187304896 cites W3047863327 @default.
- W3187304896 cites W3106318401 @default.
- W3187304896 cites W3109824650 @default.
- W3187304896 cites W3149837290 @default.
- W3187304896 cites W3171752851 @default.
- W3187304896 cites W808508909 @default.
- W3187304896 hasPublicationYear "2021" @default.
- W3187304896 type Work @default.
- W3187304896 sameAs 3187304896 @default.
- W3187304896 citedByCount "0" @default.
- W3187304896 crossrefType "posted-content" @default.
- W3187304896 hasAuthorship W3187304896A5057695979 @default.
- W3187304896 hasAuthorship W3187304896A5079375197 @default.
- W3187304896 hasAuthorship W3187304896A5083139588 @default.
- W3187304896 hasAuthorship W3187304896A5087251228 @default.
- W3187304896 hasConcept C11413529 @default.
- W3187304896 hasConcept C114614502 @default.
- W3187304896 hasConcept C119857082 @default.
- W3187304896 hasConcept C120665830 @default.
- W3187304896 hasConcept C121332964 @default.
- W3187304896 hasConcept C126255220 @default.
- W3187304896 hasConcept C12713177 @default.
- W3187304896 hasConcept C132525143 @default.
- W3187304896 hasConcept C154945302 @default.
- W3187304896 hasConcept C164226766 @default.
- W3187304896 hasConcept C177264268 @default.
- W3187304896 hasConcept C192209626 @default.
- W3187304896 hasConcept C199360897 @default.
- W3187304896 hasConcept C2776330181 @default.
- W3187304896 hasConcept C33923547 @default.
- W3187304896 hasConcept C41008148 @default.
- W3187304896 hasConcept C80444323 @default.
- W3187304896 hasConcept C97541855 @default.
- W3187304896 hasConceptScore W3187304896C11413529 @default.
- W3187304896 hasConceptScore W3187304896C114614502 @default.
- W3187304896 hasConceptScore W3187304896C119857082 @default.
- W3187304896 hasConceptScore W3187304896C120665830 @default.
- W3187304896 hasConceptScore W3187304896C121332964 @default.
- W3187304896 hasConceptScore W3187304896C126255220 @default.
- W3187304896 hasConceptScore W3187304896C12713177 @default.
- W3187304896 hasConceptScore W3187304896C132525143 @default.
- W3187304896 hasConceptScore W3187304896C154945302 @default.
- W3187304896 hasConceptScore W3187304896C164226766 @default.
- W3187304896 hasConceptScore W3187304896C177264268 @default.
- W3187304896 hasConceptScore W3187304896C192209626 @default.
- W3187304896 hasConceptScore W3187304896C199360897 @default.
- W3187304896 hasConceptScore W3187304896C2776330181 @default.
- W3187304896 hasConceptScore W3187304896C33923547 @default.
- W3187304896 hasConceptScore W3187304896C41008148 @default.
- W3187304896 hasConceptScore W3187304896C80444323 @default.
- W3187304896 hasConceptScore W3187304896C97541855 @default.
- W3187304896 hasLocation W31873048961 @default.
- W3187304896 hasOpenAccess W3187304896 @default.
- W3187304896 hasPrimaryLocation W31873048961 @default.
- W3187304896 hasRelatedWork W2095897379 @default.
- W3187304896 hasRelatedWork W2277342451 @default.
- W3187304896 hasRelatedWork W2395439732 @default.
- W3187304896 hasRelatedWork W2562068195 @default.