Commonsense Reasoning to Guide Deep Learning for Scene Understanding (Extended Abstract)

Abstract

Our architecture uses non-monotonic logical reasoning with incomplete commonsense domain knowledge, and incremental inductive learning, to guide the construction of deep network models from a small number of training examples. Experimental results in the context of a robot reasoning about the partial occlusion of objects and the stability of object configurations in simulated images indicate an improvement in reliability and a reduction in computational effort in comparison with an architecture based just on deep networks.

Cite

Text

Sridharan and Mota. "Commonsense Reasoning to Guide Deep Learning for Scene Understanding (Extended Abstract)." International Joint Conference on Artificial Intelligence, 2020. doi:10.24963/IJCAI.2020/661

Markdown

[Sridharan and Mota. "Commonsense Reasoning to Guide Deep Learning for Scene Understanding (Extended Abstract)." International Joint Conference on Artificial Intelligence, 2020.](https://mlanthology.org/ijcai/2020/sridharan2020ijcai-commonsense/) doi:10.24963/IJCAI.2020/661

BibTeX

@inproceedings{sridharan2020ijcai-commonsense,
  title     = {{Commonsense Reasoning to Guide Deep Learning for Scene Understanding (Extended Abstract)}},
  author    = {Sridharan, Mohan and Mota, Tiago},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2020},
  pages     = {4760-4764},
  doi       = {10.24963/IJCAI.2020/661},
  url       = {https://mlanthology.org/ijcai/2020/sridharan2020ijcai-commonsense/}
}