Matches in SemOpenAlex for { <https://semopenalex.org/work/W2250031007> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W2250031007 abstract "When reinforcement learning is applied to large state spaces, such as those occurring in playing board games, the use of a good function approximator to learn to approximate the value function is very important. In previous research, multi-layer perceptrons have often been quite successfully used as function approximator for learning to play particular games with temporal difference learning. With the recent developments in deep learning, it is important to study if using multiple hidden layers or particular network structures can help to improve learning the value function. In this paper, we compare five different structures of multilayer perceptrons for learning to play the game Tic-Tac-Toe 3D, both when training through self-play and when training against the same fixed opponent they are tested against. We compare three fully connected multilayer perceptrons with a different number of hidden layers and/or hidden units, as well as two structured ones. These structured multilayer perceptrons have a first hidden layer that is only sparsely connected to the input layer, and has units that correspond to the rows in Tic-Tac-Toe 3D. This allows them to more easily learn the contribution of specific patterns on the corresponding rows. One of the two structured multilayer perceptrons has a second hidden layer that is fully connected to the first one, which allows the neural network to learn to non-linearly integrate the information in these detected patterns. The results on Tic-Tac-Toe 3D show that the deep structured neural network with integrated pattern detectors has the strongest performance out of the compared multilayer perceptrons against a fixed opponent, both through self-training and through training against this fixed opponent." @default.
- W2250031007 created "2016-06-24" @default.
- W2250031007 creator A5016131956 @default.
- W2250031007 creator A5057962472 @default.
- W2250031007 creator A5060596453 @default.
- W2250031007 date "2015-12-01" @default.
- W2250031007 modified "2023-09-27" @default.
- W2250031007 title "Temporal Difference Learning for the Game Tic-Tac-Toe 3D: Applying Structure to Neural Networks" @default.
- W2250031007 cites W1515851193 @default.
- W2250031007 cites W1601359266 @default.
- W2250031007 cites W2028145673 @default.
- W2250031007 cites W2041367235 @default.
- W2250031007 cites W2076063813 @default.
- W2250031007 cites W2099001564 @default.
- W2250031007 cites W2100677568 @default.
- W2250031007 cites W2101786389 @default.
- W2250031007 cites W2101926813 @default.
- W2250031007 cites W2103196307 @default.
- W2250031007 cites W2112796928 @default.
- W2250031007 cites W2132994929 @default.
- W2250031007 cites W2145339207 @default.
- W2250031007 cites W2145934551 @default.
- W2250031007 cites W2146598871 @default.
- W2250031007 cites W2147800946 @default.
- W2250031007 cites W2149043954 @default.
- W2250031007 cites W2163605009 @default.
- W2250031007 cites W2131600418 @default.
- W2250031007 doi "https://doi.org/10.1109/ssci.2015.89" @default.
- W2250031007 hasPublicationYear "2015" @default.
- W2250031007 type Work @default.
- W2250031007 sameAs 2250031007 @default.
- W2250031007 citedByCount "8" @default.
- W2250031007 countsByYear W22500310072016 @default.
- W2250031007 countsByYear W22500310072017 @default.
- W2250031007 countsByYear W22500310072018 @default.
- W2250031007 countsByYear W22500310072019 @default.
- W2250031007 countsByYear W22500310072022 @default.
- W2250031007 crossrefType "proceedings-article" @default.
- W2250031007 hasAuthorship W2250031007A5016131956 @default.
- W2250031007 hasAuthorship W2250031007A5057962472 @default.
- W2250031007 hasAuthorship W2250031007A5060596453 @default.
- W2250031007 hasBestOaLocation W22500310072 @default.
- W2250031007 hasConcept C108583219 @default.
- W2250031007 hasConcept C119857082 @default.
- W2250031007 hasConcept C14036430 @default.
- W2250031007 hasConcept C153180895 @default.
- W2250031007 hasConcept C154945302 @default.
- W2250031007 hasConcept C178790620 @default.
- W2250031007 hasConcept C185592680 @default.
- W2250031007 hasConcept C196340769 @default.
- W2250031007 hasConcept C2779227376 @default.
- W2250031007 hasConcept C41008148 @default.
- W2250031007 hasConcept C50644808 @default.
- W2250031007 hasConcept C60908668 @default.
- W2250031007 hasConcept C78458016 @default.
- W2250031007 hasConcept C86803240 @default.
- W2250031007 hasConcept C97541855 @default.
- W2250031007 hasConceptScore W2250031007C108583219 @default.
- W2250031007 hasConceptScore W2250031007C119857082 @default.
- W2250031007 hasConceptScore W2250031007C14036430 @default.
- W2250031007 hasConceptScore W2250031007C153180895 @default.
- W2250031007 hasConceptScore W2250031007C154945302 @default.
- W2250031007 hasConceptScore W2250031007C178790620 @default.
- W2250031007 hasConceptScore W2250031007C185592680 @default.
- W2250031007 hasConceptScore W2250031007C196340769 @default.
- W2250031007 hasConceptScore W2250031007C2779227376 @default.
- W2250031007 hasConceptScore W2250031007C41008148 @default.
- W2250031007 hasConceptScore W2250031007C50644808 @default.
- W2250031007 hasConceptScore W2250031007C60908668 @default.
- W2250031007 hasConceptScore W2250031007C78458016 @default.
- W2250031007 hasConceptScore W2250031007C86803240 @default.
- W2250031007 hasConceptScore W2250031007C97541855 @default.
- W2250031007 hasLocation W22500310071 @default.
- W2250031007 hasLocation W22500310072 @default.
- W2250031007 hasLocation W22500310073 @default.
- W2250031007 hasOpenAccess W2250031007 @default.
- W2250031007 hasPrimaryLocation W22500310071 @default.
- W2250031007 hasRelatedWork W1501213224 @default.
- W2250031007 hasRelatedWork W2991591812 @default.
- W2250031007 hasRelatedWork W3196362139 @default.
- W2250031007 hasRelatedWork W3211546796 @default.
- W2250031007 hasRelatedWork W4231994957 @default.
- W2250031007 hasRelatedWork W4294067781 @default.
- W2250031007 hasRelatedWork W4322750901 @default.
- W2250031007 hasRelatedWork W4362544900 @default.
- W2250031007 hasRelatedWork W4366674482 @default.
- W2250031007 hasRelatedWork W4367364209 @default.
- W2250031007 isParatext "false" @default.
- W2250031007 isRetracted "false" @default.
- W2250031007 magId "2250031007" @default.
- W2250031007 workType "article" @default.