Ítem
Acceso Abierto

Aprendizaje por refuerzo de un parser semántico óptimo en DRT

Mostrar el registro sencillo de la publicación
dc.contributor.advisorAndrade Lotero, Édgar José
dc.creatorPiza Londoño, Jessenia
dc.creator.degreeMagíster en Matemáticas Aplicadas y Ciencias de la Computación
dc.creator.degreetypeFull time
dc.date.accessioned2024-08-13T20:32:42Z
dc.date.available2024-08-13T20:32:42Z
dc.date.created2024-08-12
dc.descriptionEste documento se trata del procesamiento de lenguaje natural (NLP, por sus siglas en inglés), que se enfoca en desarrollar sistemas de comunicación efectivos entre computadoras y humanos. Aunque los mayores avances en esta área se han logrado mediante grandes modelos de lenguaje (LLMs, por sus siglas en inglés), estos suelen ser imprecisos en dominios regidos por reglas, como las relaciones espaciales o las normas legales. Para abordar estos dominios, se utilizan parsers semánticos que asignan representaciones lógicas a los textos a través del análisis de su estructura sintáctica y la interpretación semántica. Sin embargo, estos parsers son complejos y su diseño es complicado debido a la implementación manual de reglas específicas. Este estudio propone un enfoque innovador que utiliza el aprendizaje por refuerzo profundo para desarrollar un parser semántico que pueda aprender y adaptarse automáticamente. El agente, a través de recompensas, optimizará su comportamiento con el tiempo, lo que podría tener un impacto significativo en el avance del procesamiento de lenguaje natural.
dc.description.abstractThis document is about natural language processing (NLP), which focuses on developing effective communication systems between computers and humans. While the most significant advances in this area have been achieved through large language models (LLMs), these models often lack precision in rule-governed domains, such as spatial relations or legal norms. To address these domains, semantic parsers are used to assign logical representations to texts by analyzing their syntactic structure and semantic interpretation. However, these parsers are complex, and their design is challenging due to the manual implementation of specific rules. This study proposes an innovative approach using deep reinforcement learning to develop a semantic parser that can learn and adapt automatically. Through rewards, the agent will optimize its behavior over time, which could have a significant impact on the advancement of natural language processing.
dc.format.extent44 PP
dc.format.mimetypeapplication/pdf
dc.identifier.doihttps://doi.org/10.48713/10336_43268
dc.identifier.urihttps://repository.urosario.edu.co/handle/10336/43268
dc.language.isospa
dc.publisherUniversidad del Rosariospa
dc.publisher.departmentEscuela de Ingeniería, Ciencia y Tecnologíaspa
dc.publisher.programMaestría en Matemáticas Aplicadas y Ciencias de la Computaciónspa
dc.rightsAttribution-NonCommercial-ShareAlike 4.0 International*
dc.rights.accesRightsinfo:eu-repo/semantics/openAccess
dc.rights.accesoAbierto (Texto Completo)
dc.rights.urihttp://creativecommons.org/licenses/by-nc-sa/4.0/*
dc.source.bibliographicCitationEisenstein, Jacob (2019) Introduction to Natural Language Processing. : MIT Press;
dc.source.bibliographicCitationDavidson, Donald (2001) Essays on Actions and Events: Philosophical Essays Volume 1. Oxford, GB: Clarendon Press;
dc.source.bibliographicCitationKamp, H; Reyle, U (1993) From Discourse to Logic. Dordrecht: Kluwer;
dc.source.bibliographicCitationGeurts, Bart; Beaver, David I; Maier, Emar (2020) Discourse Representation Theory. : Metaphysics Research Lab, Stanford University; Disponible en: https://plato.stanford.edu/archives/spr2020/entries/discourse-representation-theory/.
dc.source.bibliographicCitationvan Noord, Rik; Abzianidze, Lasha; Toral, Antonio; Bos, Johan (2018) Exploring Neural Methods for Parsing Discourse Representation Structures. En: Transactions of the Association for Computational Linguistics. Vol. 6; pp. 619-633 Cambridge, MA: MIT Press; Disponible en: https://aclanthology.org/Q18-1043; http://dx.doi.org/10.1162/tacl_a_00241. Disponible en: 10.1162/tacl_a_00241.
dc.source.bibliographicCitationBos, Johan (2008) Wide-Coverage Semantic Analysis with Boxer. En: Semantics in Text Processing. STEP 2008 Conference Proceedings. pp. 277-286 : College Publications; Disponible en: https://aclanthology.org/W08-2222.
dc.source.bibliographicCitationvan Lambalgen, Michiel; Hamm, Fritz (2008) The Proper Treatment of Events. En: Explorations in Semantics.: Wiley; 9780470759226;
dc.source.bibliographicCitationAndrade-Lotero, Edgar (2006) Meaning and Form in Event Calculus. MSc. Thesis. : ILLC, Universiteit van Amsterdam;
dc.source.bibliographicCitationJurafsky, Daniel; Martin, James (2008) Speech and Language Processing. : Prentice Hall;
dc.source.bibliographicCitation van Harmelen, Frank; Lifschitz, Vladimir; Porter, Bruce (2008) Handbook of Knowledge Representation. Amsterdam: Elsevier;
dc.source.bibliographicCitationVaswani, Ashish; Shazeer, Noam; Parmar, Niki; Uszkoreit, Jakob; Jones, Llion; Gomez, Aidan N; Kaiser, Lukasz; Polosukhin, Illia (2023) Attention Is All You Need. En: arXiv [cs.CL]. Disponible en: http://arxiv.org/abs/1706.03762.
dc.source.bibliographicCitationTraylor, Aaron; Feiman, Roman; Pavlick, Ellie; Zong, Chengqing; Xia, Fei; Li, Wenjie; Navigli, Roberto (2021) AND does not mean OR: Using Formal Languages to Study Language Models'. En: Proceedings of the 59th Annual Meeting of the Association for. pp. 158-167 : Association for Computational Linguistics;
dc.source.bibliographicCitationKassner, Nora; Krojer, Benno; Schütze, Hinrich; Fernández, Raquel; Linzen, Tal (2020) Are Pretrained Language Models Symbolic Reasoners over Knowledge?. En: Proceedings of the 24th Conference on Computational Natural Language. pp. 552-564 : Association for Computational Linguistics;
dc.source.bibliographicCitationBasmov, Victoria; Goldberg, Yoav; Tsarfaty, Reut (2024) Simple Linguistic Inferences of Large Language Models (LLMs): Blind Spots. En: arXiv [cs.CL]. Disponible en: http://arxiv.org/abs/2305.14785.
dc.source.bibliographicCitationMinsky, Marvin; Winston, P H (1975) A framework for representing knowledge. En: The Psychology of Computer Vision.: McGraw-Hill;
dc.source.bibliographicCitationMcDermott, D V (1987) A critique of pure reason. En: Computational Intelligence. Vol. 3; pp. 151-160
dc.source.bibliographicCitationGamut, L T F (1991) Logic, Language and Meaning Vol. 2. : University of Chicago Press;
dc.source.bibliographicCitationMueller, Erik T (2006) Common Sense Reasoning. : Elsevier;
dc.source.bibliographicCitationSutton, Richard S; Barto, Andrew G (2018) Reinforcement Learning. : MIT Press;
dc.source.bibliographicCitationMnih, Volodymyr; Kavukcuoglu, Koray; Silver, David; Graves, Alex; Antonoglou, Ioannis; Wierstra, Daan; Riedmiller, Martin A (2013) Playing Atari with Deep Reinforcement Learning. En: CoRR. Vol. abs/1312.5602;
dc.source.bibliographicCitationAbzianidze, Lasha; Bjerva, Johannes; Evang, Kilian; Haagsma, Hessel; van Noord, Rik; Ludmann, Pierre; Nguyen, Duc-Duy; Bos, Johan (2017) The Parallel Meaning Bank: Towards a Multilingual Corpus of Translations. En: Proceedings of the 15th Conference of the European Chapter of the. pp. 242-247 : Association for Computational Linguistics; Disponible en: https://aclanthology.org/E17-2039.
dc.source.bibliographicCitationZai, Alexander; Brown, Brandon (2020) Deep Reinforcement Learning in Action. : Manning Publications;
dc.source.bibliographicCitationOzdemir, Sinan (2023) Quick Start Guide to Large Language Models: Strategies and Best Practices. : Addison-Wesley Professional;
dc.source.bibliographicCitationMontague, Richard (1974) Formal Philosophy: Selected Papers of Richard Montague. : New Haven: Yale University Press;
dc.source.bibliographicCitationHugging Face (2024) sentence-transformers/distiluse-base-multilingual-cased-v1. Disponible en: https://huggingface.co/sentence-transformers/distiluse-base-multilingual-cased-v1.
dc.source.bibliographicCitationMnih, Volodymyr; Kavukcuoglu, Koray; Silver, David; Others, (2015) Human-level control through deep reinforcement learning. En: Nature. Vol. 518; No. 7540; pp. 529-533 : Nature Publishing Group;
dc.source.bibliographicCitationBinz, Marcel; Schulz, Eric (2023) Using cognitive psychology to understand GPT-3. En: Proceedings of the National Academy of Sciences. Vol. 120; No. 6; Proceedings of the National Academy of Sciences; 1091-6490; Disponible en: http://dx.doi.org/10.1073/pnas.2218523120. Disponible en: 10.1073/pnas.2218523120.
dc.source.bibliographicCitationUc-Cetina, Víctor; Navarro-Guerrero, Nicolás; Martin-Gonzalez, Anabel; Weber, Cornelius; Wermter, Stefan (2022) Survey on reinforcement learning for language processing. En: Artificial Intelligence Review. Vol. 56; No. 2; pp. 1543–1575 : Springer Science and Business Media LLC; 1573-7462; Disponible en: http://dx.doi.org/10.1007/s10462-022-10205-5. Disponible en: 10.1007/s10462-022-10205-5.
dc.source.bibliographicCitationSchulman, John; Wolski, Filip; Dhariwal, Prafulla; Radford, Alec; Klimov, Oleg (2017) Proximal Policy Optimization Algorithms. En: arXiv [cs.LG]. Disponible en: http://arxiv.org/abs/1707.06347.
dc.source.bibliographicCitationStarc, Janez; Mladenić, Dunja (2016) Joint learning of ontology and semantic parser from text. En: arXiv [cs.AI]. Disponible en: http://arxiv.org/abs/1601.00901.
dc.source.bibliographicCitationYang, Zhilin; Qi, Peng; Zhang, Saizheng; Bengio, Yoshua; Cohen, William W; Salakhutdinov, Ruslan; Manning, Christopher D (2018) HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering. En: arXiv [cs.CL]. Disponible en: http://arxiv.org/abs/1809.09600.
dc.source.bibliographicCitationGeva, Mor; Khashabi, Daniel; Segal, Elad; Khot, Tushar; Roth, Dan; Berant, Jonathan (2021) Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit. En: Transactions of the Association for Computational Linguistics. Vol. 9; pp. 346-361 Disponible en: https://api.semanticscholar.org/CorpusID:230799347.
dc.source.bibliographicCitationHo, Xanh; Duong Nguyen, Anh-Khoa; Sugawara, Saku; Aizawa, Akiko; Scott, Donia; Bel, Nuria; Zong, Chengqing (2020) Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of. En: Proceedings of the 28th International Conference on Computational. pp. 6609-6625 : International Committee on Computational Linguistics; Disponible en: https://aclanthology.org/2020.coling-main.580; http://dx.doi.org/10.18653/v1/2020.coling-main.580. Disponible en: 10.18653/v1/2020.coling-main.580.
dc.source.instnameinstname:Universidad del Rosario
dc.source.reponamereponame:Repositorio Institucional EdocURspa
dc.subjectRepresentación formal del lenguaje
dc.subjectRazonamiento automático
dc.subjectInferencia Lógica
dc.subjectTeoría de la Representación del Discurso
dc.subjectProcesamiento de Lenguaje Natural
dc.subject.keywordFormal representation of language
dc.subject.keywordAutomatic reasoning
dc.subject.keywordLogical inference
dc.subject.keywordDiscourse representation theory
dc.subject.keywordNatural language processing
dc.titleAprendizaje por refuerzo de un parser semántico óptimo en DRT
dc.title.TranslatedTitleReinforcement Learning of an Optimal Semantic Parser in DRT
dc.typebachelorThesis
dc.type.documentTrabajo de grado
dc.type.spaTrabajo de grado
local.department.reportEscuela de Ingeniería, Ciencia y Tecnología
local.regionesBogotá
Archivos
Bloque original
Mostrando1 - 2 de 2
Miniatura
Nombre:
Parser_semantico_Piza_Londoño_Jessenia.pdf
Tamaño:
1.1 MB
Formato:
Adobe Portable Document Format
Descripción:
Descargar
Miniatura
Nombre:
Referencias_Parser_semantico_Piza_Londoño_Jessenia.ris
Tamaño:
10.01 KB
Formato:
Descripción:
Descargar
Colecciones
Maestría en Matemáticas Aplicadas y Ciencias de la Computación