2026-09-19T16:01:04Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1301842026-06-16T18:15:31Zcom_1721.1_7582com_1721.1_7581col_1721.1_131022
Joshua B. Tenenbaum and Armando Solar-Lezama.
Ellis, Kevin,Ph. D.(Kevin M.)Massachusetts Institute of Technology.
Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences.
Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences
2021-03-22T17:05:57Z
2021-03-22T17:05:57Z
2020
2020
https://hdl.handle.net/1721.1/130184
1241081869
Thesis: Ph. D. in Cognitive Science, Massachusetts Institute of Technology, Department of Brain and Cognitive Sciences, September, 2020
Cataloged from student-submitted PDF version of thesis.
Includes bibliographical references (pages 213-224).
The future of machine learning should have a knowledge representation that supports, at a minimum, several features: Expressivity, interpretability, the potential for reuse by both humans and machines, while also enabling sample-efficient generalization. Here we argue that programs-i.e., source code-are a knowledge representation which can contribute to the project of capturing these elements of intelligence. This research direction however requires new program synthesis algorithms which can induce programs solving a range of AI tasks. This program induction challenge confronts two primary obstacles: the space of all programs is infinite, so we need a strong inductive bias or prior to steer us toward the correct programs; and even if we have that prior, effectively searching through the vast combinatorial space of all programs is generally intractable. We introduce algorithms that learn to induce programs, with the goal of addressing these two primary obstacles. Focusing on case studies in vision, computational linguistics, and learning-to-learn, we develop an algorithmic toolkit for learning inductive biases over programs as well as learning to search for programs, drawing on probabilistic, neural, and symbolic methods. Together this toolkit suggests ways in which program induction can contribute to AI, and how we can use learning to improve program synthesis technologies.
by Kevin Ellis.
Ph.D. in Cognitive Science
Ph.D.inCognitiveScience Massachusetts Institute of Technology, Department of Brain and Cognitive Sciences
224 pages
eng
Massachusetts Institute of Technology
MIT theses may be protected by copyright. Please reuse MIT thesis content according to the MIT Libraries Permissions Policy, which is available through the URL provided.
http://dspace.mit.edu/handle/1721.1/7582
Brain and Cognitive Sciences.
Algorithms for learning to induce programs
Thesis
application/pdf
2021-03-22T17:05:27Z
Publication
Doctoral
Brain
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="2333caad-db1b-4516-8080-94bf5bb7b9ff">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Language>eng</Language>
<Title>Algorithms for learning to induce programs</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2020</PublicationDate>
<Authors>
<Author>
<DisplayName>Ellis, Kevin,Ph. D.(Kevin M.)Massachusetts Institute of Technology.</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://dspace.mit.edu/handle/1721.1/7582</License>
<Keyword>Brain and Cognitive Sciences.</Keyword>
<Abstract>The future of machine learning should have a knowledge representation that supports, at a minimum, several features: Expressivity, interpretability, the potential for reuse by both humans and machines, while also enabling sample-efficient generalization. Here we argue that programs-i.e., source code-are a knowledge representation which can contribute to the project of capturing these elements of intelligence. This research direction however requires new program synthesis algorithms which can induce programs solving a range of AI tasks. This program induction challenge confronts two primary obstacles: the space of all programs is infinite, so we need a strong inductive bias or prior to steer us toward the correct programs; and even if we have that prior, effectively searching through the vast combinatorial space of all programs is generally intractable. We introduce algorithms that learn to induce programs, with the goal of addressing these two primary obstacles. Focusing on case studies in vision, computational linguistics, and learning-to-learn, we develop an algorithmic toolkit for learning inductive biases over programs as well as learning to search for programs, drawing on probabilistic, neural, and symbolic methods. Together this toolkit suggests ways in which program induction can contribute to AI, and how we can use learning to improve program synthesis technologies.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>