Elamur Research⌕
☰
home / papers / 2015 / Commonsense Reasoning and Commonsense Knowledge in Artificial Intelligence

Commonsense Reasoning and Commonsense Knowledge in Artificial Intelligence

Paper · 2015 · Ernest Davis and Gary Marcus

Summary

This is a survey article on commonsense reasoning and knowledge about the everyday world, one of the oldest problems in artificial intelligence and one where progress has been slowest. Using examples, the authors show where such knowledge is lacking. In NLP and machine translation, ambiguity cannot be resolved without it. In computer vision, partially visible or occluded objects cannot be recognized without it. Robots need it to respond appropriately to unforeseen events. The authors then describe four areas where well-developed theory already exists: reasoning about taxonomies, time, actions and change, and qualitative reasoning based on the signs of quantities (sign calculus). Next, they explain why the general problem is so difficult and critically examine the main approaches: large knowledge bases such as CYC, automatic extraction of facts from the web, and crowdsourcing. The main conclusion is that there is no universal solution: each direction has its advantages, but none provides reliable reasoning. The authors call for creating benchmarks and metrics, systematically evaluating CYC, combining approaches, covering broader kinds of reasoning, and drawing on cognitive science. For NLP projects, the article is useful as an argument about the limits of purely statistical methods.

Key points

  • Winograd schemas. These are pairs of sentences in which the referent of a pronoun changes when a single word is replaced (the example with the city council and the demonstrators: "feared" vs. "advocated"). Solving them requires knowledge about the world rather than linguistic cues. A collection of such examples was compiled by Levesque, Davis, and Morgenstern (2012).
  • The Google Translate test (9.06.2015, English → German). In short sentences about an electrician and a telephone, "working" is translated correctly (arbeitet / funktioniert). When a few words are inserted between the key words, however, the choice of sense becomes wrong. The statistical substitute for common sense breaks down on more complex sentences.
  • Watson (Jeopardy). According to Kalyanpur (2012), taxonomic, geographic, and temporal reasoning made only a small contribution to its success. The system is based on highly tuned rules for matching words to fragments of documents such as Wikipedia.
  • Taxonomies (WordNet, semantic networks) support transitivity, property inheritance, and default inheritance: Robin inherits Bird → CanFly, but Penguin does not. For cross-cutting or abstract categories they become problematic.
  • Web mining and crowdsourcing. The OIE system does not distinguish "wood" as a material from "wood" as a forest. The crowdsourced ConceptNet 3.5 contains contradictory links, uneven frequency of facts, and polysemous nodes (cook as a profession and as an action), and the restriction to binary relations limits its expressiveness. At the same time, for the query "What is made of wood?" (November 2014) it gave mostly sensible answers.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓New York UniversityNew York · United States219 works · 103,487 citations

 Fields, topics and concepts

📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 works

🧭 Topics 1Research subjects and languages the work is about.Reasoning · 1,685 works

💡 Concepts 1Key ideas and methods explained on our concept pages.Artificial intelligence · 49 works

Cite

Cite as Ernest Davis and Gary Marcus. 2015. Commonsense Reasoning and Commonsense Knowledge in Artificial Intelligence. Communications of the ACM 58(9):92–103.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.1145/2701413

@inproceedings{davis2015commonsense,
  title = {{Commonsense Reasoning and Commonsense Knowledge in Artificial Intelligence}},
  author = {Ernest Davis and Gary Marcus},
  booktitle = {Communications of the ACM 58(9):92–103},
  year = {2015},
  doi = {10.1145/2701413},
  url = {https://doi.org/10.1145/2701413}
}

References

49 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Allison, B., Guthrie, D., & Guthrie, L. (2006). Another Look at the Data Sparsity Problem. Text, Speech, and Dialogue: 9th International Conference, (pp. 327-334). find
  2. Baader, F., Horrocks, I., & Sattler, U. (2008). Descriptions Logics. In F. Van Harmelen, V. Lifschitz, & B. Porter, Handbook of Knowledge Representation (pp. 135-179). Amsterdam: Elsevier. find
  3. Bar-Hillel, Y. (1960). The present status of automatic translation of languges. In F. Alt, Advances in Computers (Vol. 1, pp. 91-163). New York: Academic Press. find
  4. Bojanowski, P., Lajugie, R., Bach, F., Laptev, I., Ponce, J., Schmid, C., & Sivic, J. (2014). Weakly supervised action labeling in videos under ordering constraints. ECCV, (pp. 628-643). find
  5. Brewka, G., Niemelli, I., & Truszczynski, M. (2008). Nonmonotonic Reasoning. In F. Van Harmelen, V. Lifschitz, & B. Porter, Handbook of Knowledge Representation (pp. 239-284). Amsterdam: Elsevier. find
  6. Cohn, A., & Renz, J. (2007). Qualitative Spatial Reasoning. In F. van Harmelen, V. Lifschitz, & B. Porter, Handbook of Knowledge Representation (pp. 551-596). Elsevier. find
  7. Conesa, J., Storey, V., & Sugumaran, V. (2008). Improving web-query processing through semantic knowledge. Data and Knowledge Engineering, 66(1), 18-34. find
  8. Conesa, J., Storey, V., & Sugumaran, V. (2010). Usability of upper level ontologies: The case of ResearchCyc. Data and Knowledge Engineering, 69, 343-356. find
  9. Curtis, J., Matthews, G., & Baxter, D. (2005). On the effective use of Cyc in a question answering system. IJCAI Workshop on Knowledge and Reasoning for Answering Questions. find
  10. Davis, E. (1998). The naive physics perplex. AI Magazine, 19(4), 51-79. From http://www.aaai.org/ojs/index.php/aimagazine/article/view/1424/1323 de Kleer, J. (1975). Qualitative and quantitative knowledge in classical mechanics. ↗
  11. MIT AI Lab. de Kleer, J., & Brown, J. (1985). A qualitative physics based on confluences. In D. Bobrow, Qualitative Reasoning about Physical Systems (pp. 7-84). Cambridge: MIT Press. find
  12. Dennett, D. (2013). Intuition Pumps and Other Tools for Thinking. Norton. find
  13. Etizoni, O., Fader, A., Christensen, J., Soderland, S., & Mausam. (2011). Open Information Extraction: The Second Generation. IJAI, (pp. 3-10). find
  14. Etzioni, O., Cafarella, M., Downey, D., Kok, S., Popsecu, A., Shaked, T., . . . Yates, A. (2004). Web-scale extraction in KnowItAll (preliminary results). 13th International Conference on World Wide Web, (pp. 100-110). Retrieved from http://dl.acm.org/citation.cfm?id=988687 ↗
  15. Ferrein, A., Fritz, C., & Lakemeyer, G. (2005). Using Golog for Deliberation and Team Coordination in Robotic Soccer. KI (Kuntzliche Intelligenz), 19(1), 24-30. find
  16. Fisher, M. (2008). Temporal Representation and Reasoning. In F. Van Harmelen, V. Lifschitz, & B. Porter, Handbook of Knowledge Representation (pp. 513-550). Amsterdam: Elsevier. find
  17. Forbus, K. (1985). Qualitative process theory. In D. Bobrow, Qualitative Reasoning about Physical Systems (pp. 85-168). Cambridge, Mass.: MIT Press. find
  18. Forbus, K., Birnbaum, L., Wagner, E., Baker, J., & Witbrock, M. (2005). Analogy, Intelligent IR, and Knowledge Integration for Intelligence Analysis: Situation Tracking and the Whodunit Problem. International Conference on Intelligence Analysis. find
  19. Fromherz, M., Bobrow, D., & de Kleer, J. (2003). Model-based Computing for Design and Control of Reconfigurable Systems. AI Magazine, 24(4), 120. find
  20. Gene Ontology Consortium. (2004). The Gene Ontology (GO) database and informatics resource. Nucleic acids research, 32(suppl. 1), D258-D261. find
  21. Gentner, D., & Forbus, K. (2011). Computational models of analogy. WIREs Cognitive Science, 2, 266-276. find
  22. Halpern, J. (2003). Reasoning about Uncertainty. Cambridge, MA: MIT Press. find
  23. Havasi, C., Pustjekovsky, J., Speer, R., & Lieberman, H. (2009). Digital Intuition: Applying Common Sense Using Dimensionality Reduction. IEEE Intelligent Systems, 24(4), 24-35. doi:dx.doi.org/10.1109/MIS.2009.72 — [doi:10.1109/mis.2009.72](https://doi.org/10.1109/mis.2009.72) find
  24. Hearst, M. (1992). Automatic acquisition of hyponyms from large text corpora. ACL, (pp. 539-545).
  25. Hofstadter, D., & Sander, E. (2013). Surfaces and Essences: Analogy as the Fuel and Fire of Thinking. New York: Basic Books. find
  26. Kalyanpur, A. (2012). Structured data and inference in DeepQA. IBM Journal of Research and Development, 53(3-4), 10:1-14. find
  27. Kolodner, J. (1993). Case-based reasoning. San Mateo, Calif.: Morgan Kaufmann. find
  28. Kuehne, S. (2004). Understanding natural language description of physical phenomena. Evanston, I.: Northwestern University. find
  29. Kuipers, B. (1986). Qualitative simulation. Artificial Intelligence, 29, 289-338. find
  30. Lenat, D., Prakash, M., & Shepherd, M. (1985). CYC: Using common sense knowledge to overcome brittleness and knowledge acquisition bottlenecks. AI Magazine, 6(4), 65-85. find
  31. Levesque, H., Davis, E., & Morgenstern, L. (2012). The Winograd schema challenge. Principles of Knowledge Representation and Reasoing. find
  32. Lovett, A., Tomei, E., Forbus, K., & Usher, J. (2009). Solving Geometric Analogy Problems through Two-Stage Analogical Mapping. Cognitive Science, 33(7), 1192-1231. find
  33. Miller, G. (1995). WordNet:A Lexical Database for English. Communications of the ACM, 38(11), 39-41.
  34. Minsky, M. (1975). A Framework for Representing Knowledge. In P. Winston, The Psychology of Computer Vision. New York: McGraw Hill. find
  35. Mitchell, T., Cohen, W., Hruschka, E., Talukdar, P., Betteridge, J., & Carlson, A. (2015). Never Ending Learning. AAAI. find
  36. Mueller, E. (2006). Commonsense Reasoning. San Francisco: Morgan Kaufmann. find
  37. Pearl, J. (1988). Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. San Mateo, CA: Morgan Kaufmann. find
  38. Peppas, P. (2008). Belief Revision. In F. Van Harmelen, V. Lifschitz, & B. Porter, Handbook of Knowledge Representation (pp. 317-359). Amsterdam: Elsevier. find
  39. Pisanelli, D. (2004). Ontologies in Medicine. Amsterdam: IOS Press. find
  40. Price, C., Pugh, D., Wilson, M., & Snooke, N. (1995). The FLAME System: Automating electrical failure mode and effects analysis (FEMA). IEEE Reliability and Maintainability Symposium, (pp. 90-95). find
  41. Reiter, R. (2001). Knowledge in Action: Logical Foundations for Specifying and Implementing Dynamical Systemsi. Cambridge, Mass.: MIT Press. find
  42. Russell, B., Torralba, A., Murphy, K., & Freeman, W. (2008). Labelme: a database and web-based tool for image annotation. International Journal of Computer Vision, 77(1-3), 157-173. find
  43. Schank, R., & Abelson, R. (1977). Scripts, Plans, Goals, and Understanding. Hillsdale, N.J.: Lawrence Erlbaum. find
  44. Schubert, L. (2015). Semantic Representation. AAAI. find
  45. Shepard, B., Matuszek, C., Fraser, C., Wechtenhiser, W., Crabbe, D., Gungordu, Z. J., . . . Larson, E. (2005). A knowledge-base approach to network security: Applying Cyc in the domain of network risk assessment. Association for the Advancement of Artificial Intelligence, (pp. 1563-1568). find
  46. Surdeanu, M. (2013). Overview of the tac2013 knowledge base population evaluation: English slot filling and temporal slot filling. Proceedings of the Sixth Text Analysis Conference (TAC 2013). find
  47. Winograd, T. (1972). Understanding Natural Language. New York: Academic Press. find
  48. Woods, W. (1975). What's in a Link: Foundations for Semantic Networks. In D. Bobrow, & A. Collins, Representation and Understanding: Studies in Cognitive Science. New York: Academic Press. find
  49. Wu, W., Li, H., Wang, H., & Zhu, K. Q. (2012). Probase: A probabilistic taxonomy for text understanding. ACM SIGMOD. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/davis-2015-commonsense-reasoning-commonsense-knowledge-artificial/