Matches in SemOpenAlex for { <https://semopenalex.org/work/W2912943342> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W2912943342 endingPage "261" @default.
- W2912943342 startingPage "258" @default.
- W2912943342 abstract "Deep Reinforcement Learning (DRL) is poised to revolutionize the field of artificial intelligence (AI) and represents a crucial step towards building autonomous systems with a higher-level understanding of the world around them. In particular, deep reinforcement learning has changed the landscape of autonomous agents by achieving superhuman performance on board game Go, a significant milestone in AI research. In this project, we attempt to train a Deep RL network on Demon Attack – an Atari 2600 game and test the model on different game environments to investigate the feasibility of applying Transfer Learning on environments with same action space but slightly different state space. We further extend the project to use established Reinforcement Learning techniques such as DQN, Dueling DQN, and SARSA to examine whether RL agents can be generalized on unfamiliar environments by fine-tuning the hyperparameters. Finally, we borrow classic regularization techniques like 2 regularization and dropout from the world of supervised learning and probe whether these techniques which have received very limited attention in the domain of reinforcement learning are effective in reducing overfitting of Deep RL networks. Deep Networks are expensive to train and complex models take weeks to train using expensive GPUs. We find that the use of the above techniques prevents the network from overfitting on the current environment and gives satisfactory results when tested on slightly different environments thus enabling substantial savings in training time & resources." @default.
- W2912943342 created "2019-02-21" @default.
- W2912943342 creator A5023923022 @default.
- W2912943342 creator A5069537581 @default.
- W2912943342 creator A5078481539 @default.
- W2912943342 creator A5083235215 @default.
- W2912943342 date "2019-04-02" @default.
- W2912943342 modified "2023-09-26" @default.
- W2912943342 title "Improving generalization in reinforcement learning on Atari 2600 games" @default.
- W2912943342 hasPublicationYear "2019" @default.
- W2912943342 type Work @default.
- W2912943342 sameAs 2912943342 @default.
- W2912943342 citedByCount "0" @default.
- W2912943342 crossrefType "journal-article" @default.
- W2912943342 hasAuthorship W2912943342A5023923022 @default.
- W2912943342 hasAuthorship W2912943342A5069537581 @default.
- W2912943342 hasAuthorship W2912943342A5078481539 @default.
- W2912943342 hasAuthorship W2912943342A5083235215 @default.
- W2912943342 hasConcept C108583219 @default.
- W2912943342 hasConcept C119857082 @default.
- W2912943342 hasConcept C150899416 @default.
- W2912943342 hasConcept C154945302 @default.
- W2912943342 hasConcept C22019652 @default.
- W2912943342 hasConcept C2776135515 @default.
- W2912943342 hasConcept C41008148 @default.
- W2912943342 hasConcept C50644808 @default.
- W2912943342 hasConcept C8642999 @default.
- W2912943342 hasConcept C97541855 @default.
- W2912943342 hasConceptScore W2912943342C108583219 @default.
- W2912943342 hasConceptScore W2912943342C119857082 @default.
- W2912943342 hasConceptScore W2912943342C150899416 @default.
- W2912943342 hasConceptScore W2912943342C154945302 @default.
- W2912943342 hasConceptScore W2912943342C22019652 @default.
- W2912943342 hasConceptScore W2912943342C2776135515 @default.
- W2912943342 hasConceptScore W2912943342C41008148 @default.
- W2912943342 hasConceptScore W2912943342C50644808 @default.
- W2912943342 hasConceptScore W2912943342C8642999 @default.
- W2912943342 hasConceptScore W2912943342C97541855 @default.
- W2912943342 hasIssue "1" @default.
- W2912943342 hasLocation W29129433421 @default.
- W2912943342 hasOpenAccess W2912943342 @default.
- W2912943342 hasPrimaryLocation W29129433421 @default.
- W2912943342 hasRelatedWork W2555488107 @default.
- W2912943342 hasRelatedWork W2588283865 @default.
- W2912943342 hasRelatedWork W2883899184 @default.
- W2912943342 hasRelatedWork W2893662673 @default.
- W2912943342 hasRelatedWork W2896183040 @default.
- W2912943342 hasRelatedWork W2922485250 @default.
- W2912943342 hasRelatedWork W2944710665 @default.
- W2912943342 hasRelatedWork W2997633275 @default.
- W2912943342 hasRelatedWork W3008102108 @default.
- W2912943342 hasRelatedWork W3013640302 @default.
- W2912943342 hasRelatedWork W3016488666 @default.
- W2912943342 hasRelatedWork W3018569081 @default.
- W2912943342 hasRelatedWork W3097894767 @default.
- W2912943342 hasRelatedWork W3103078407 @default.
- W2912943342 hasRelatedWork W3111442723 @default.
- W2912943342 hasRelatedWork W3136181012 @default.
- W2912943342 hasRelatedWork W3171711942 @default.
- W2912943342 hasRelatedWork W3206615896 @default.
- W2912943342 hasRelatedWork W3210166519 @default.
- W2912943342 hasRelatedWork W3213159396 @default.
- W2912943342 hasVolume "5" @default.
- W2912943342 isParatext "false" @default.
- W2912943342 isRetracted "false" @default.
- W2912943342 magId "2912943342" @default.
- W2912943342 workType "article" @default.