{"id":{"repo_id":"nus","oai_identifier":"oai:scholarbank.nus.edu.sg:10635/237301"},"canonical_url":"https://search.dev.ndltd.org/etd/nus/oai:scholarbank.nus.edu.sg:10635/237301","repository":{"repo_id":"nus","name":"National University of Singapore","base_url":"https://scholarbank.nus.edu.sg/oai/request"},"display":{"title":"ENHANCING DEEP LEARNING WITH SYMBOLIC DOMAIN KNOWLEDGE","abstract":"Deep neural networks have brought significant advances to various tasks in machine learning and artificial intelligence. However, despite their effectiveness and flexibility, deep models have two drawbacks: low data efficiency and a lack of robustness. Firstly, deep neural networks often require large amounts of training data. On the other hand, symbolic domain knowledge is often available in addition to data. The first part of this thesis aims to improve data efficiency by incorporating symbolic domain knowledge. We propose logic graph embedding frameworks, Logic Embedding Network with Semantic Regularization (LENSR) and Temporal-Logic Embedded Automata Framework (T-LEAF), which take propositional logic and linear temporal logic as inputs, respectively. Secondly, recent work has shown that deep neural networks are vulnerable to adversarial attacks. In the second part of the thesis, we propose to leverage prior knowledge to defend against adversarial attacks in RL settings using the Knowledge-based Policy Recycling (KPR) framework.","abstract_html":"Deep neural networks have brought significant advances to various tasks in machine learning and artificial intelligence. However, despite their effectiveness and flexibility, deep models have two drawbacks: low data efficiency and a lack of robustness. Firstly, deep neural networks often require large amounts of training data. On the other hand, symbolic domain knowledge is often available in addition to data. The first part of this thesis aims to improve data efficiency by incorporating symbolic domain knowledge. We propose logic graph embedding frameworks, Logic Embedding Network with Semantic Regularization (LENSR) and Temporal-Logic Embedded Automata Framework (T-LEAF), which take propositional logic and linear temporal logic as inputs, respectively. Secondly, recent work has shown that deep neural networks are vulnerable to adversarial attacks. In the second part of the thesis, we propose to leverage prior knowledge to defend against adversarial attacks in RL settings using the Knowledge-based Policy Recycling (KPR) framework.","abstract_has_math":false,"creators":["XIE YAQI"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-10-31","date_published":"2022-10-31","updated_at":"2026-07-24T03:31:26Z","subjects":["Artificial Intelligence, Machine Learning, Knowledge Representation, Reinforcement Learning, Knowledge Embedding, Graph Neural Networks"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["XIE YAQI"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.issued","label":"Date","values":["2022-10-31"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://scholarbank.nus.edu.sg/handle/10635/237301"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Artificial Intelligence, Machine Learning, Knowledge Representation, Reinforcement Learning, Knowledge Embedding, Graph Neural Networks"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://scholarbank.nus.edu.sg/bitstreams/9dfd3185-993a-4308-a9c3-1b9ea34b4971/download"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Deep neural networks have brought significant advances to various tasks in machine learning and artificial intelligence. However, despite their effectiveness and flexibility, deep models have two drawbacks: low data efficiency and a lack of robustness. Firstly, deep neural networks often require large amounts of training data. On the other hand, symbolic domain knowledge is often available in addition to data. The first part of this thesis aims to improve data efficiency by incorporating symbolic domain knowledge. We propose logic graph embedding frameworks, Logic Embedding Network with Semantic Regularization (LENSR) and Temporal-Logic Embedded Automata Framework (T-LEAF), which take propositional logic and linear temporal logic as inputs, respectively. Secondly, recent work has shown that deep neural networks are vulnerable to adversarial attacks. In the second part of the thesis, we propose to leverage prior knowledge to defend against adversarial attacks in RL settings using the Knowledge-based Policy Recycling (KPR) framework."]},{"key":"dc:format.checksum.md5","label":"Dc Format Checksum Md5","values":["6fbea0a970be7d80c636560c8e3dc47f","5286d3f65c28dfb5ab2ebb2857fe33ac"]},{"key":"dc:title","label":"Title","values":["ENHANCING DEEP LEARNING WITH SYMBOLIC DOMAIN KNOWLEDGE"]}]}],"canonical_facts":{"dc:creator":["XIE YAQI"],"dc:date.issued":["2022-10-31"],"dc:description.abstract":["Deep neural networks have brought significant advances to various tasks in machine learning and artificial intelligence. However, despite their effectiveness and flexibility, deep models have two drawbacks: low data efficiency and a lack of robustness. Firstly, deep neural networks often require large amounts of training data. On the other hand, symbolic domain knowledge is often available in addition to data. The first part of this thesis aims to improve data efficiency by incorporating symbolic domain knowledge. We propose logic graph embedding frameworks, Logic Embedding Network with Semantic Regularization (LENSR) and Temporal-Logic Embedded Automata Framework (T-LEAF), which take propositional logic and linear temporal logic as inputs, respectively. Secondly, recent work has shown that deep neural networks are vulnerable to adversarial attacks. In the second part of the thesis, we propose to leverage prior knowledge to defend against adversarial attacks in RL settings using the Knowledge-based Policy Recycling (KPR) framework."],"dc:format.checksum.md5":["6fbea0a970be7d80c636560c8e3dc47f","5286d3f65c28dfb5ab2ebb2857fe33ac"],"dc:identifier.uri":["https://scholarbank.nus.edu.sg/bitstreams/9dfd3185-993a-4308-a9c3-1b9ea34b4971/download"],"dc:relation.isreferencedby":["https://scholarbank.nus.edu.sg/handle/10635/237301"],"dc:subject":["Artificial Intelligence, Machine Learning, Knowledge Representation, Reinforcement Learning, Knowledge Embedding, Graph Neural Networks"],"dc:title":["ENHANCING DEEP LEARNING WITH SYMBOLIC DOMAIN KNOWLEDGE"],"dc:type":["Thesis"]},"updated_at":"2026-07-24T03:31:26Z"}