FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3192022969> ?p ?o ?g. }

• W3192022969 endingPage "1" @default.
• W3192022969 startingPage "1" @default.
• W3192022969 abstract "We study the problem of balancing effectiveness and efficiency in automated feature selection. Feature selection is to find an optimal feature subset from large feature space. After exploring many feature selection methods, we observe a computational dilemma: 1) traditional feature selection (e.g., mRMR) is mostly efficient, but difficult to identify the best subset; 2) the emerging reinforced feature selection automatically navigates feature space to search the best subset, but is usually inefficient. Are automation and efficiency always apart from each other? Can we bridge the gap between effectiveness and efficiency under automation? Motivated by this dilemma, we aim to develop a novel feature space navigation method. In our preliminary work, we leveraged interactive reinforcement learning to accelerate feature selection by external trainer-agent interaction. Our preliminary work can be significantly improved by modeling the structured knowledge of its downstream task (e.g., decision tree) as learning feedback. In this journal version, we propose a novel interactive and closed-loop architecture to simultaneously model interactive reinforcement learning (IRL) and decision tree feedback (DTF). Specifically, IRL is to create an interactive feature selection loop and DTF is to feed structured feature knowledge back to the loop. The DTF improves IRL from two aspects. First, the tree-structured feature hierarchy generated by decision tree is leveraged to improve state representation. In particular, we represent the selected feature subset as an undirected graph of feature-feature correlations and a directed tree of decision features. We propose a new embedding method capable of empowering Graph Convolutional Network (GCN) to jointly learn state representation from both the graph and the tree. Second, the tree-structured feature hierarchy is exploited to develop a new reward scheme. In particular, we personalize reward assignment of agents based on decision tree feature importance. In addition, observing agents’ actions can also be a feedback, we devise another new reward scheme, to weigh and assign reward based on the selected frequency ratio of each agent in historical action records. Finally, we present extensive experiments with real-world datasets to demonstrate the improved performances of our method." @default.
• W3192022969 created "2021-08-16" @default.
• W3192022969 creator A5032187620 @default.
• W3192022969 creator A5038836690 @default.
• W3192022969 creator A5050609316 @default.
• W3192022969 creator A5067731925 @default.
• W3192022969 creator A5088664989 @default.
• W3192022969 creator A5091745851 @default.
• W3192022969 date "2021-01-01" @default.
• W3192022969 modified "2023-10-15" @default.
• W3192022969 title "Interactive Reinforcement Learning for Feature Selection with Decision Tree in the Loop" @default.
• W3192022969 cites W1523989055 @default.
• W3192022969 cites W1574447377 @default.
• W3192022969 cites W1969685488 @default.
• W3192022969 cites W1975777241 @default.
• W3192022969 cites W1983801113 @default.
• W3192022969 cites W2017337590 @default.
• W3192022969 cites W2040584032 @default.
• W3192022969 cites W2053744708 @default.
• W3192022969 cites W2092829070 @default.
• W3192022969 cites W2117423905 @default.
• W3192022969 cites W2125212374 @default.
• W3192022969 cites W2135046866 @default.
• W3192022969 cites W2141559645 @default.
• W3192022969 cites W2142222368 @default.
• W3192022969 cites W2145339207 @default.
• W3192022969 cites W2151170651 @default.
• W3192022969 cites W2154053567 @default.
• W3192022969 cites W2600702321 @default.
• W3192022969 cites W2792217087 @default.
• W3192022969 cites W2799784543 @default.
• W3192022969 cites W2905510914 @default.
• W3192022969 cites W2953043480 @default.
• W3192022969 cites W2963432546 @default.
• W3192022969 cites W2963653811 @default.
• W3192022969 cites W3081189998 @default.
• W3192022969 cites W3126977517 @default.
• W3192022969 cites W3127251420 @default.
• W3192022969 doi "https://doi.org/10.1109/tkde.2021.3102120" @default.
• W3192022969 hasPublicationYear "2021" @default.
• W3192022969 type Work @default.
• W3192022969 sameAs 3192022969 @default.
• W3192022969 citedByCount "5" @default.
• W3192022969 countsByYear W31920229692022 @default.
• W3192022969 countsByYear W31920229692023 @default.
• W3192022969 crossrefType "journal-article" @default.
• W3192022969 hasAuthorship W3192022969A5032187620 @default.
• W3192022969 hasAuthorship W3192022969A5038836690 @default.
• W3192022969 hasAuthorship W3192022969A5050609316 @default.
• W3192022969 hasAuthorship W3192022969A5067731925 @default.
• W3192022969 hasAuthorship W3192022969A5088664989 @default.
• W3192022969 hasAuthorship W3192022969A5091745851 @default.
• W3192022969 hasBestOaLocation W31920229692 @default.
• W3192022969 hasConcept C119857082 @default.
• W3192022969 hasConcept C138885662 @default.
• W3192022969 hasConcept C148483581 @default.
• W3192022969 hasConcept C154945302 @default.
• W3192022969 hasConcept C2776401178 @default.
• W3192022969 hasConcept C41008148 @default.
• W3192022969 hasConcept C41895202 @default.
• W3192022969 hasConcept C59404180 @default.
• W3192022969 hasConcept C83665646 @default.
• W3192022969 hasConcept C84525736 @default.
• W3192022969 hasConcept C97541855 @default.
• W3192022969 hasConceptScore W3192022969C119857082 @default.
• W3192022969 hasConceptScore W3192022969C138885662 @default.
• W3192022969 hasConceptScore W3192022969C148483581 @default.
• W3192022969 hasConceptScore W3192022969C154945302 @default.
• W3192022969 hasConceptScore W3192022969C2776401178 @default.
• W3192022969 hasConceptScore W3192022969C41008148 @default.
• W3192022969 hasConceptScore W3192022969C41895202 @default.
• W3192022969 hasConceptScore W3192022969C59404180 @default.
• W3192022969 hasConceptScore W3192022969C83665646 @default.
• W3192022969 hasConceptScore W3192022969C84525736 @default.
• W3192022969 hasConceptScore W3192022969C97541855 @default.
• W3192022969 hasFunder F4320306076 @default.
• W3192022969 hasLocation W31920229691 @default.
• W3192022969 hasLocation W31920229692 @default.
• W3192022969 hasOpenAccess W3192022969 @default.
• W3192022969 hasPrimaryLocation W31920229691 @default.
• W3192022969 hasRelatedWork W2104602707 @default.
• W3192022969 hasRelatedWork W2977487495 @default.
• W3192022969 hasRelatedWork W3087493185 @default.
• W3192022969 hasRelatedWork W3210877509 @default.
• W3192022969 hasRelatedWork W3215669537 @default.
• W3192022969 hasRelatedWork W4295514622 @default.
• W3192022969 hasRelatedWork W4300428003 @default.
• W3192022969 hasRelatedWork W4309118558 @default.
• W3192022969 hasRelatedWork W4318348373 @default.
• W3192022969 hasRelatedWork W4366376591 @default.
• W3192022969 isParatext "false" @default.
• W3192022969 isRetracted "false" @default.
• W3192022969 magId "3192022969" @default.
• W3192022969 workType "article" @default.