Reasoning, nonmonotonicity and learning in connectionist networks that capture propositional knowledge

This paper deals with a criticism of Ryle's claim that the so called Intellectualist legend leads to an infinite regress. Critics have attempted to show that Ryle's argument cannot even get off the ground since its two basic premises cannot be true at the same time. In the paper I argue that this objection is based on a misinterpretation of Ryle's argumentation, which is complex and consists of two arguments, not of a single one as it is claimed. One of Ryle's argument attacks the thesis that an intelligent act is an indirect result of propositional knowledge, while the other, which I call the Asymmetry argument, claims that not every manifestation of knowledge that is accompanied with the manifestation of knowing how. In the paper I argue that both Ryle's arguments are valid and resistant to recent critique so it can be said that Ryle's distinction between knowledge that and knowing how is still an important distinction within contemporary epistemology.

Download Full-text

Conclusion

10.1093/oso/9780190852603.003.0008 ◽

2018 ◽

Author(s):

Harold D. Morales

Keyword(s):

Identity Politics ◽

Practical Knowledge ◽

Propositional Knowledge ◽

History Of ◽

Muslim Communities ◽

Struggle For Recognition

The conclusion provides a summary of key developments in the history of Latino Muslim communities and also critically explores future possibilities. While weaving a trail among the history of Islamic Spain, the Alianza Islamica, and subsequent Latino Muslim organizations, the struggle for recognition through solidarity groups emerges as a prominent theme throughout the book. However, this approach to liberation raises complex issues regarding the efficacy and logics of identity politics. Drawing on various sources, I argue that practical knowledge of how to know and how to be in relation with one another may circumvent identity politics premised on static propositional knowledge of groups like Latino Muslims.

Download Full-text

Theology and the Knowledge of Persons

10.1093/oso/9780198796732.003.0010 ◽

2017 ◽

Cited By ~ 1

Author(s):

Eleonore Stump

Keyword(s):

Propositional Knowledge ◽

Ways Of Thinking ◽

Way Of Thinking ◽

The Difference ◽

Knowledge Of Persons

This chapter is concerned with the question of the difference between philosophy and theology. It rejects certain prevalent ways of thinking about this difference. It argues that a more promising way of thinking about these disciplines is to be found in their names: “philosophy” in its etymology means something like the love of wisdom; “theology” in its etymology means something like the word with regard to God. God, unlike wisdom, is not an abstract universal. Rather, and by virtue of being characterized by mind and will, God is more nearly a person. The chapter spells out the implications of this difference, arguing that the knowledge at issue when we do theology is irreducible to propositional knowledge. Rather, it is a knowledge of persons. The chapter illuminates the role of the knowledge of persons in theological discussion and draws some conclusions about the methodology which will be useful to theology.

Download Full-text

Conclusions and Further Tasks

10.1093/oso/9780199693658.003.0013 ◽

2018 ◽

Author(s):

Catherine Rowett

Keyword(s):

Conceptual Knowledge ◽

Propositional Knowledge ◽

Philosophical Method ◽

Type Iii

The chapter starts by telling a narrative to explain how and why the author came to reject the mistaken assumptions with which the research began, and how these initial assumptions had assumed false dichotomies familiar from existing work in the field. The chapter thereby explains why the results presented in Chapters 1–12 might seem unexpected. It draws together the chief philosophical lessons of those chapters, highlighting the fact that Plato is right about (i) how conceptual knowledge differs from both propositional knowledge and recognition of tokens, (ii) the different sense of ‘being’ involved in knowing ‘what it is’, about a type, (iii) the value of images and icons in the philosophical method, and (iv) the irrelevance of Socratic definitions and other bogus criteria for knowledge. Finally, it sketches some possible ways in which a further volume might apply the results to other dialogues.

Download Full-text

Using Noise to Compute Error Surfaces in Connectionist Networks: A Novel Means of Reducing Catastrophic Forgetting

Neural Computation ◽

10.1162/08997660260028700 ◽

2002 ◽

Vol 14 (7) ◽

pp. 1755-1769 ◽

Cited By ~ 6

Author(s):

Robert M. French ◽

Nick Chater

Keyword(s):

Feedforward Networks ◽

First Derivative ◽

Error Surface ◽

Connectionist Networks ◽

New Information ◽

Catastrophic Interference ◽

Derivative Information

In error-driven distributed feedforward networks, new information typically interferes, sometimes severely, with previously learned information. We show how noise can be used to approximate the error surface of previously learned information. By combining this approximated error surface with the error surface associated with the new information to be learned, the network's retention of previously learned items can be improved and catastrophic interference significantly reduced. Further, we show that the noise-generated error surface is produced using only first-derivative information and without recourse to any explicit error information.

Download Full-text