Temporal and Object Quantification Networks

TitleTemporal and Object Quantification Networks
Publication TypeConference Paper
Year of Publication2021
AuthorsMao, J, Luo, Z, Gan, C, Tenenbaum, JB, Wu, J, Kaelbling, LPack, Ullman, TD
EditorZhou, Z-H
Tertiary AuthorsGini, M
Conference NameThirtieth International Joint Conference on Artificial Intelligence {IJCAI-21}Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence
Date Published06/2021
Conference LocationMontreal, Canada
Abstract

We present Temporal and Object Quantification Networks (TOQ-Nets), a new class of neuro-symbolic networks with a structural bias that enables them to learn to recognize complex relational-temporal events. This is done by including reasoning layers that implement finite-domain quantification over objects and time. The structure allows them to generalize directly to input instances with varying numbers of objects in temporal sequences of varying lengths. We evaluate TOQ-Nets on input domains that require recognizing event-types in terms of complex temporal relational patterns. We demonstrate that TOQ-Nets can generalize from small amounts of data to scenarios containing more objects than were present during training and to temporal warpings of input sequences.

URLhttps://www.ijcai.org/proceedings/2021
DOI10.24963/ijcai.2021/386
Download:  PDF icon 0386.pdf

CBMM Relationship: 

  • CBMM Funded