2026-09-19T06:54:25Zhttps://dspace.mit.edu/server/oai/request

oai:dspace.mit.edu:1721.1/1218102026-06-16T18:52:44Zcom_1721.1_7582com_1721.1_7581col_1721.1_131022

Tomaso Poggio. Lewis, Owen,Ph. D.Massachusetts Institute of Technology. Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences. Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences 2019-07-18T20:31:33Z 2019-07-18T20:31:33Z 2019 2019 https://hdl.handle.net/1721.1/121810 1103712241 Thesis: Ph. D., Massachusetts Institute of Technology, Department of Brain and Cognitive Sciences, 2019 Cataloged from PDF version of thesis. Includes bibliographical references (pages 91-100). Neural networks and probabilistic models have different and in many ways complementary strengths and weaknesses: neural networks are flexible and support efficient inference, but rely on large quantities of labeled training data. Probabilistic models can learn from fewer examples, but in many cases remain limited by time-consuming inference algorithms. Thus, both classes of models have drawbacks that both limit their engineering applications and prevent them from being fully satisfying as process models of human learning. This thesis aims to address this state of affairs from both directions, exploring case studies where we make neural networks that learn from less data, and in which we design more efficient inference procedures for generative models. First, we explore recurrent neural networks that learn list-processing procedures (sort, reverse, etc.), and show how ideas from type theory and programming language theory can be used to design a data augmentation scheme that enables effective learning from small datasets. Next, we show how error-driven proposal mechanisms can speed up stochastic search for generative model inversion, first developing a symbolic model for inferring Boolean functions and Horn clause theories, and then a general-purpose neural network model for doing inference in continuous domains such as inverse graphics. by Owen Lewis. Ph.D. Ph.D. Massachusetts Institute of Technology, Department of Brain and Cognitive Sciences 100 pages eng Massachusetts Institute of Technology MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission. http://dspace.mit.edu/handle/1721.1/7582 Brain and Cognitive Sciences. Structured learning and inference with neural networks and generative models Thesis application/pdf 2019-07-18T20:31:30Z Publication Doctoral Brain unknown unknown <Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="163711c5-748f-4bd3-9c57-165f80dcbc7f"> <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>Structured learning and inference with neural networks and generative models</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2019</PublicationDate> <Authors> <Author> <DisplayName>Lewis, Owen,Ph. D.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>Neural networks and probabilistic models have different and in many ways complementary strengths and weaknesses: neural networks are flexible and support efficient inference, but rely on large quantities of labeled training data. Probabilistic models can learn from fewer examples, but in many cases remain limited by time-consuming inference algorithms. Thus, both classes of models have drawbacks that both limit their engineering applications and prevent them from being fully satisfying as process models of human learning. This thesis aims to address this state of affairs from both directions, exploring case studies where we make neural networks that learn from less data, and in which we design more efficient inference procedures for generative models. First, we explore recurrent neural networks that learn list-processing procedures (sort, reverse, etc.), and show how ideas from type theory and programming language theory can be used to design a data augmentation scheme that enables effective learning from small datasets. Next, we show how error-driven proposal mechanisms can speed up stochastic search for generative model inversion, first developing a symbolic model for inferring Boolean functions and Horn clause theories, and then a general-purpose neural network model for doing inference in continuous domains such as inverse graphics.</Abstract> <Access xmlns="http://purl.org/coar/access_right" > </Access> </Publication>