2026-09-18T23:58:11Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1432492022-06-16T03:04:31Zcom_1721.1_7582com_1721.1_7581col_1721.1_131022
Solar-Lezama, Armando
Inala, Jeevana Priya
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2022-06-15T13:07:14Z
2022-06-15T13:07:14Z
2022-02
2022-03-04T20:48:02.548Z
https://hdl.handle.net/1721.1/143249
https://orcid.org/0000-0003-1843-589X
This thesis shows that looking at intelligent systems through the lens of neurosymbolic models has several benefits over traditional deep learning approaches. Neurosymbolic models contain symbolic programmatic constructs such as loops and conditionals and continuous neural components. The symbolic part makes the model interpretable, generalizable, and robust, while the neural part handles the complexity of the intelligent systems. Concretely, this thesis presents two classes of neurosymbolic models—state-machines and neurosymbolic transformers and evaluates them on two case studies—reinforcement-learning based autonomous systems and multirobot systems. These case studies showed that the learned neurosymbolic models are human-readable, can be extrapolated to unseen scenarios, and can handle robust objectives in the specification. To efficiently learn these neurosymbolic models, we introduce neurosymbolic learning algorithms that leverage the latest techniques from machine learning and program synthesis.
Ph.D.
Massachusetts Institute of Technology
In Copyright - Educational Use Permitted
Copyright MIT
http://rightsstatements.org/page/InC-EDU/1.0/
Neurosymbolic Learning for Robust and Reliable Intelligent Systems
Thesis
application/pdf
Doctoral
Doctor of Philosophy
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="e1cbc3c0-843c-4d73-860c-e7dea6c8184f">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Title>Neurosymbolic Learning for Robust and Reliable Intelligent Systems</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2022-02</PublicationDate>
<Authors>
<Author>
<DisplayName>Inala, Jeevana Priya</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://rightsstatements.org/page/InC-EDU/1.0/</License>
<Abstract>This thesis shows that looking at intelligent systems through the lens of neurosymbolic models has several benefits over traditional deep learning approaches. Neurosymbolic models contain symbolic programmatic constructs such as loops and conditionals and continuous neural components. The symbolic part makes the model interpretable, generalizable, and robust, while the neural part handles the complexity of the intelligent systems. Concretely, this thesis presents two classes of neurosymbolic models—state-machines and neurosymbolic transformers and evaluates them on two case studies—reinforcement-learning based autonomous systems and multirobot systems. These case studies showed that the learned neurosymbolic models are human-readable, can be extrapolated to unseen scenarios, and can handle robust objectives in the specification. To efficiently learn these neurosymbolic models, we introduce neurosymbolic learning algorithms that leverage the latest techniques from machine learning and program synthesis.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>