Matches in SemOpenAlex for { <https://semopenalex.org/work/W3032255069> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W3032255069 abstract "The development of a satisfying and rigorous mathematical understanding of the performance of neural networks is a major challenge in artificial intelligence. Against this background, we study the expressive power of neural networks through the example of the classical NP-hard knapsack problem. Our main contribution is a class of recurrent neural networks (RNNs) with rectified linear units that are iteratively applied to each item of a knapsack instance and thereby compute optimal or provably good solution values. We show that an RNN of depth four and width depending quadratically on the profit of an optimum knapsack solution is sufficient to find optimum knapsack solutions. We also prove the following tradeoff between the size of an RNN and the quality of the computed knapsack solution: for knapsack instances consisting of n items, an RNN of depth five and width w computes a solution of value at least [Formula: see text] times the optimum solution value. Our results build on a classical dynamic programming formulation of the knapsack problem and a careful rounding of profit values that are also at the core of the well-known fully polynomial-time approximation scheme for the knapsack problem. A carefully conducted computational study qualitatively supports our theoretical size bounds. Finally, we point out that our results can be generalized to many other combinatorial optimization problems that admit dynamic programming solution methods, such as various shortest path problems, the longest common subsequence problem, and the traveling salesperson problem. History: Andrea Lodi, Area Editor for Design & Analysis of Algorithms–Discrete. Funding: This work was supported by the Deutsche Forschungsgemeinschaft [Grants DFG-GRK 2434 and EXC-2046/1, Project 390685689] and the H2020 European Research Council [ScaleOpt-757481]." @default.
- W3032255069 created "2020-06-05" @default.
- W3032255069 creator A5055693054 @default.
- W3032255069 creator A5066485339 @default.
- W3032255069 date "2023-05-04" @default.
- W3032255069 modified "2023-09-30" @default.
- W3032255069 title "Provably Good Solutions to the Knapsack Problem via Neural Networks of Bounded Size" @default.
- W3032255069 cites W1597286183 @default.
- W3032255069 cites W1969186119 @default.
- W3032255069 cites W1971483538 @default.
- W3032255069 cites W1980025264 @default.
- W3032255069 cites W2059324105 @default.
- W3032255069 cites W2117666205 @default.
- W3032255069 cites W2136658137 @default.
- W3032255069 cites W2143612262 @default.
- W3032255069 cites W2401610261 @default.
- W3032255069 cites W2528305538 @default.
- W3032255069 cites W2626188202 @default.
- W3032255069 cites W2919115771 @default.
- W3032255069 cites W3047863327 @default.
- W3032255069 cites W3099132019 @default.
- W3032255069 cites W3160912907 @default.
- W3032255069 cites W3173557077 @default.
- W3032255069 cites W4205585032 @default.
- W3032255069 cites W4236127247 @default.
- W3032255069 cites W4236362309 @default.
- W3032255069 cites W4287180320 @default.
- W3032255069 cites W4298227433 @default.
- W3032255069 doi "https://doi.org/10.1287/ijoc.2021.0225" @default.
- W3032255069 hasPublicationYear "2023" @default.
- W3032255069 type Work @default.
- W3032255069 sameAs 3032255069 @default.
- W3032255069 citedByCount "2" @default.
- W3032255069 countsByYear W30322550692021 @default.
- W3032255069 countsByYear W30322550692023 @default.
- W3032255069 crossrefType "journal-article" @default.
- W3032255069 hasAuthorship W3032255069A5055693054 @default.
- W3032255069 hasAuthorship W3032255069A5066485339 @default.
- W3032255069 hasConcept C111919701 @default.
- W3032255069 hasConcept C113138325 @default.
- W3032255069 hasConcept C126255220 @default.
- W3032255069 hasConcept C136625980 @default.
- W3032255069 hasConcept C137836250 @default.
- W3032255069 hasConcept C154945302 @default.
- W3032255069 hasConcept C17722475 @default.
- W3032255069 hasConcept C194146004 @default.
- W3032255069 hasConcept C33923547 @default.
- W3032255069 hasConcept C37307619 @default.
- W3032255069 hasConcept C37404715 @default.
- W3032255069 hasConcept C41008148 @default.
- W3032255069 hasConcept C50644808 @default.
- W3032255069 hasConcept C52692508 @default.
- W3032255069 hasConcept C94569963 @default.
- W3032255069 hasConcept C99140742 @default.
- W3032255069 hasConceptScore W3032255069C111919701 @default.
- W3032255069 hasConceptScore W3032255069C113138325 @default.
- W3032255069 hasConceptScore W3032255069C126255220 @default.
- W3032255069 hasConceptScore W3032255069C136625980 @default.
- W3032255069 hasConceptScore W3032255069C137836250 @default.
- W3032255069 hasConceptScore W3032255069C154945302 @default.
- W3032255069 hasConceptScore W3032255069C17722475 @default.
- W3032255069 hasConceptScore W3032255069C194146004 @default.
- W3032255069 hasConceptScore W3032255069C33923547 @default.
- W3032255069 hasConceptScore W3032255069C37307619 @default.
- W3032255069 hasConceptScore W3032255069C37404715 @default.
- W3032255069 hasConceptScore W3032255069C41008148 @default.
- W3032255069 hasConceptScore W3032255069C50644808 @default.
- W3032255069 hasConceptScore W3032255069C52692508 @default.
- W3032255069 hasConceptScore W3032255069C94569963 @default.
- W3032255069 hasConceptScore W3032255069C99140742 @default.
- W3032255069 hasLocation W30322550691 @default.
- W3032255069 hasOpenAccess W3032255069 @default.
- W3032255069 hasPrimaryLocation W30322550691 @default.
- W3032255069 hasRelatedWork W179363847 @default.
- W3032255069 hasRelatedWork W1966986652 @default.
- W3032255069 hasRelatedWork W2087209118 @default.
- W3032255069 hasRelatedWork W2889442766 @default.
- W3032255069 hasRelatedWork W2993365626 @default.
- W3032255069 hasRelatedWork W3032255069 @default.
- W3032255069 hasRelatedWork W3173557077 @default.
- W3032255069 hasRelatedWork W4281741337 @default.
- W3032255069 hasRelatedWork W4287774600 @default.
- W3032255069 hasRelatedWork W4365514833 @default.
- W3032255069 isParatext "false" @default.
- W3032255069 isRetracted "false" @default.
- W3032255069 magId "3032255069" @default.
- W3032255069 workType "article" @default.