Sociology Index

Commonsense Reasoning

Commonsense reasoning is also referred to as mundane reasoning. Commonsense reasoning is a term used by ethnomethodologists, derived from Alfred Schutz (1899-1959), referring to the practical or everyday reasoning used by members of society to create and sustain a sense of social reality as being objective, factual, predictable and external to themselves. Since the objectivity of the world as a practical accomplishment is the focus of ethnomethodology this kind of reasoning is a primary topic of investigation.

Commonsense Reasoning Books

Commonsense Reasoning by Erik T Mueller
Central to the idea of Artificial Intelligence is getting computers to understand simple facts about people and everyday lifewhat we call Common Sense. Amid the technical discussions about inference algorithms and knowledge representation, a larger question arises: What have we actually learned in the past 30 years about how to put Commonsense knowledge in computers? Look no further than Erik Mueller's Commonsense Reasoning for a deep and insightful survey of the state of the art in this topic. The strength of this book is that it uses a uniform representation formalism, the event calculus, to solve a variety of commonsense reasoning problems.

Commonsense Reasoning Abstracts

Learning in Order to Reason: The Approach. SOFSEM: Theory and Practice of Informatics (1996), D. Roth
Abstract: Any theory aimed at understanding commonsense reasoning, the process that humans use to cope with the mundane but complex aspects of the world in evaluating everyday situations, should account for its flexibility, its adaptability, and the speed with which it is performed. Current theories of reasoning, however, do not satisfy these requirements, a fact we attribute, at least partly, to their separation from learning. 

An architecture of diversity for commonsense reasoning, IBM Systems Journal, vol. 41(3). Mccarthy, J., Marvin, M., Sloman, A., Gong, L., Lau, T., Morgenstern, L., Mueller, E.T., Riecken, D., Singh, M. and Singh, P. 
Abstract: This paper discusses commonsense reasoning and what makes it difficult for computers. The paper contends that commonsense reasoning is too hard a problem to solve using any single artificial intelligence technique. A multilevel architecture is proposed that consists of diverse reasoning and representation techniques that collaborate and reflect in order to allow the best techniques to be used for the many situations that arise in commonsense reasoning. Story understanding, specifically, understanding and answering questions about progressively harder children's texts, is presented as a task for evaluating and scaling up a commonsense reasoning system.

Commonsense Reasoning and Commonsense Knowledge in Artificial Intelligence.
Ernest Davis, Dept. of Computer Science, New York University
Gary Marcus, Dept. of Psychology, New York University
Abstract Since the earliest days of artificial intelligence, it has been recognized that commonsense reasoning is one of the central challenges in the field. However, progress in this area has on the whole been frustratingly slow. In this review paper, we discuss why commonsense reasoning is needed to achieve human-level performance in tasks like natural language processing, vision, and robotics, why the problem is so difficult, and why progress has been slow. We also discuss four particular areas where substantial progress has been made, the techniques that have been attempted, and prospects for going forward.

A Simple Method for Commonsense Reasoning
TrieuH.Trinh, QuocV.Le
Abstract: Commonsense reasoning is a long-standing challenge for deep learning. For example, it is dif´Čücult to use neural networks to tackle the Winograd Schema dataset. In this paper, we present a simple method for commonsense reasoning with neural networks, using unsupervised learning. Key to our method is the use of language models, trained on a massive amount of unlableddata, to score multiple choice questions posed by commonsense reasoning tests. On both Pronoun Disambiguation and Winograd Schema challenges, our models outperform previous state-of-the-art methods by a large margin, without using expensive annotated knowledge bases or hand-engineered features. Analysis also shows that our system successfully discovers important features of the context that decide the correct answer, indicating a good grasp of commonsense knowledge.