Temporal and Object Quantification Networks
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.
Cite
Text
Mao et al. "Temporal and Object Quantification Networks." International Joint Conference on Artificial Intelligence, 2021. doi:10.24963/IJCAI.2021/386Markdown
[Mao et al. "Temporal and Object Quantification Networks." International Joint Conference on Artificial Intelligence, 2021.](https://mlanthology.org/ijcai/2021/mao2021ijcai-temporal/) doi:10.24963/IJCAI.2021/386BibTeX
@inproceedings{mao2021ijcai-temporal,
title = {{Temporal and Object Quantification Networks}},
author = {Mao, Jiayuan and Luo, Zhezheng and Gan, Chuang and Tenenbaum, Joshua B. and Wu, Jiajun and Kaelbling, Leslie Pack and Ullman, Tomer D.},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2021},
pages = {2804-2811},
doi = {10.24963/IJCAI.2021/386},
url = {https://mlanthology.org/ijcai/2021/mao2021ijcai-temporal/}
}