Search

Our goal is to develop models that allow a robot to efficiently understand or “ground” natural language instructions in the context of its world representation. Contemporary approaches estimate correspondences between language instructions and possible groundings such as objects, regions, and goals for actions that the robot should execute. However, these approaches typically reason in relatively small domains and do not model abstract spatial concepts such as as “rows,” “columns,” or “groups” of objects and, hence, are unable to interpret an instruction such as “pick up the middle block in the row of five blocks.” In this paper, we introduce two new models for efficient natural language understanding of robot instructions. The first model, which we call the adaptive distributed correspondence graph (ADCG), is a probabilistic model for interpreting abstract concepts that require hierarchical reasoning over constituent concrete entities as well as notions of cardinality and ordinality. Abstract grounding variables form a Markov boundary over concrete groundings, effectively de-correlating them from the remaining variables in the graph. This structure reduces the complexity of model training and inference. Inference in the model is posed as an approximate search procedure that orders factor computation such that the estimated probable concrete groundings focus the search for abstract concepts towards likely hypothesis, pruning away improbable portions of the exponentially large space of abstractions. Further, we address the issue of scalability to complex domains and introduce a hierarchical extension to a second model termed the hierarchical adaptive distributed correspondence graph (HADCG). The model utilizes the abstractions in the ADCG but infers a coarse symbolic structure from the utterance and the environment model and then performs fine-grained inference over the reduced graphical model, further improving the efficiency of inference. Empirical evaluation demonstrates accurate grounding of abstract concepts embedded in complex natural language instructions commanding a robotic torso and a mobile robot. Further, the proposed approximate inference method allows significant efficiency gains compared with the baseline, with minimal trade-off in accuracy. [ABSTRACT FROM AUTHOR]

What are the relative contributions of language and physical practice to practical understanding? The resolution of a series of puzzles depends upon the answer. I argue that language is, and must be, more central than physical practice in individual acquisition of practical understanding. Only this makes it possible for there to be a sociology of scientific knowledge, for there to be scientific specialities, for there to be a division of labour in society and for there to be a society that is more than a set of narrow and isolated worlds. Physical practice remains central to human culture but its influence is at the collective level at which languages are formed, rather than the individual level at which practical abilities are acquired. Domain languages ‘contain’ practices, and it is from these that individuals draw much, usually most, of their practical understanding. Because the individual level and the domain level have not previously been distinguished, certain philosophical problems have been wrongly cast and mistakes have been made. Domains of practice/language are embedded within one another in fractal-like relationships, and this is how we can make sense of higher levels of coordinated action. The ideas of ‘special interactional expert’, ‘practice language’ and ‘methodological interactionalism’ are introduced. [ABSTRACT FROM PUBLISHER]

Abstract: Claude Lévi-Strauss hinted that society was like a language. Kenneth Pike's distinction between emic and etic vernaculars probes the relation between the language in a society and the language of social science. Our thesis is that etic scholarly language that classifies social reality is synonymous with some deep linguistic structure (in Noam Chomsky's sense) of everyday emic language, in some measure unconscious for the user, but discoverable by science. Adding insights of Charles Morris and Charles Stevenson, we maintain that language becomes differentiated into descriptive, evaluative and prescriptive usages, each of which contains executive and emotive components. Such a language can codify societal orders, represent riches, summarize knowledge, embody beauty, define sacredness and express virtues. It can be subject to the operations of the common language of sciences, i.e. logics, mathematics and statistics. It opens an easy door to social struc- turation and can rephrase dramas of human life. [ABSTRACT FROM AUTHOR]

The article focuses on recovering relativity. The implications of relativistic approach constitute not only a favourable heuristic approach for the sociology of science itself, but also force a further reconsideration of the role which science has played as a model for practice in the human sciences. Sociologists are interested in the processes of transformation, though it is a commonplace of commentary upon the state of sociology that is in the rudimentary stage in the development of man's ability to characterize and explain them. This article is concerned with the role of relativism in furthering this enterprise, after indicating moral priorities, seeks to emphasize the positive advantages of this relativism. It also examines certain arguments which are thought to expose disqualifying flaws in the relativist position and finds them to be insignificant within the context of what are, after all, specifically sociological concerns. Socialization has a large effect on most scientific beliefs and perceptions. A relativistic frame of mind is the most appropriate for the pursuit of the sociology of science.

The article discusses several topics featured in articles in this issue. The commentary discusses the growth of moral philosophy and the impact it had on sociology in Nordic countries. In the first issue of "Acta Sociologica," co-founder Torgny Segerstedt highlighted the "Uppsala School of Sociology" and went on to outline the theoretical model of the school. Sociologists have continued to expand on the ideas of the "Uppsala School." Scholar Hans Zetterberg has continued this tradition and has expanded it to argue that language is the basis of human society.