Novel Primitive Decompositions for Real-World Physical Reasoning

Mackie Zhou, Bridget Duah, Jamie C. Macbeth
Proceedings of the Third International Workshop on Self-Supervised Learning, PMLR 192:22-34, 2022.

Abstract

In this work, we are concerned with developing cognitive representations that may enhance the ability for self-supervised learning systems to learn language as part of their world explorations. We apply insights from in-depth language understanding systems to the problem, specifically representations which decompose language inputs into language-free structures that are complex combinations of primitives representing cognitive abstractions such as object permanence, movement, and spatial relationships. These decompositions, performed by a system traditionally called a conceptual analyzer, link words with complex non-linguistic structures that engender the rich relations between language expressions and world exploration that are a familiar aspect of intelligence. We focus on improving and extending both the Conceptual Dependency (CD) representation system, its primitive decompositions, and its conceptual analyzer, choosing as our corpus the ProPara (“Process Paragraphs”) dataset, which consists of paragraphs describing biological, chemical, and physical processes of the kind that appear in grade-school science textbooks (e.g., photosynthesis, erosion). In doing so, we avoid the significant challenges of decomposing concepts involving communication, thought, and complex social interactions. To meet the challenges of this dataset, we contribute a mental motion pictures representation system with important innovations, such as using image schemas in place of CD primitives and decoupling containment relationships into separate primitives.

Cite this Paper

BibTeX
@InProceedings{pmlr-v192-zhou22a, title = {Novel Primitive Decompositions for Real-World Physical Reasoning}, author = {Zhou, Mackie and Duah, Bridget and Macbeth, Jamie C.}, booktitle = {Proceedings of the Third International Workshop on Self-Supervised Learning}, pages = {22--34}, year = {2022}, editor = {Thórisson, Kristinn R.}, volume = {192}, series = {Proceedings of Machine Learning Research}, month = {28--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v192/zhou22a/zhou22a.pdf}, url = {https://proceedings.mlr.press/v192/zhou22a.html}, abstract = {In this work, we are concerned with developing cognitive representations that may enhance the ability for self-supervised learning systems to learn language as part of their world explorations. We apply insights from in-depth language understanding systems to the problem, specifically representations which decompose language inputs into language-free structures that are complex combinations of primitives representing cognitive abstractions such as object permanence, movement, and spatial relationships. These decompositions, performed by a system traditionally called a conceptual analyzer, link words with complex non-linguistic structures that engender the rich relations between language expressions and world exploration that are a familiar aspect of intelligence. We focus on improving and extending both the Conceptual Dependency (CD) representation system, its primitive decompositions, and its conceptual analyzer, choosing as our corpus the ProPara (“Process Paragraphs”) dataset, which consists of paragraphs describing biological, chemical, and physical processes of the kind that appear in grade-school science textbooks (e.g., photosynthesis, erosion). In doing so, we avoid the significant challenges of decomposing concepts involving communication, thought, and complex social interactions. To meet the challenges of this dataset, we contribute a mental motion pictures representation system with important innovations, such as using image schemas in place of CD primitives and decoupling containment relationships into separate primitives.} }
Endnote
%0 Conference Paper %T Novel Primitive Decompositions for Real-World Physical Reasoning %A Mackie Zhou %A Bridget Duah %A Jamie C. Macbeth %B Proceedings of the Third International Workshop on Self-Supervised Learning %C Proceedings of Machine Learning Research %D 2022 %E Kristinn R. Thórisson %F pmlr-v192-zhou22a %I PMLR %P 22--34 %U https://proceedings.mlr.press/v192/zhou22a.html %V 192 %X In this work, we are concerned with developing cognitive representations that may enhance the ability for self-supervised learning systems to learn language as part of their world explorations. We apply insights from in-depth language understanding systems to the problem, specifically representations which decompose language inputs into language-free structures that are complex combinations of primitives representing cognitive abstractions such as object permanence, movement, and spatial relationships. These decompositions, performed by a system traditionally called a conceptual analyzer, link words with complex non-linguistic structures that engender the rich relations between language expressions and world exploration that are a familiar aspect of intelligence. We focus on improving and extending both the Conceptual Dependency (CD) representation system, its primitive decompositions, and its conceptual analyzer, choosing as our corpus the ProPara (“Process Paragraphs”) dataset, which consists of paragraphs describing biological, chemical, and physical processes of the kind that appear in grade-school science textbooks (e.g., photosynthesis, erosion). In doing so, we avoid the significant challenges of decomposing concepts involving communication, thought, and complex social interactions. To meet the challenges of this dataset, we contribute a mental motion pictures representation system with important innovations, such as using image schemas in place of CD primitives and decoupling containment relationships into separate primitives.
APA
Zhou, M., Duah, B. & Macbeth, J.C.. (2022). Novel Primitive Decompositions for Real-World Physical Reasoning. Proceedings of the Third International Workshop on Self-Supervised Learning, in Proceedings of Machine Learning Research 192:22-34 Available from https://proceedings.mlr.press/v192/zhou22a.html.

Related Material