China
Africa
Explore chapters and articles related to this topic
Volcano: Enabling Correctness by Design
Published in Richard Zurawski, Networked Embedded Systems, 2017
In Volcano there are three types of signals: Integer signals. These represent unsigned numbers and are of a static size between 1 and 16 bits. So, for example, a 16 bit signal can store integers in the range 0–65,535.Boolean signals. These represent truth conditions (true/false). Note that this is not the same as a 1 bit integer signal (which stores the integer values 0 or 1).Byte signals. These represent data with no Volcano-defined structure. A byte signal consists of a fixed number between 1 and 8 bytes.
Cognitively Viable Computational Models of Linguistic Knowledge
Published in Shalom Lappin, Deep Learning and Linguistic Representation, 2020
The classical formal semantic program (Davidson, 1967; Montague, 1974) seeks a recursive definition of a truth predicate which entails appropriate truth conditions for each declarative sentence in a language. To the extent that it is successful, a generalised multi-modal MT model would achieve the core part of this program. It would specify suitable correspondences between sentences and sets of situations that the sentences describe. These correspondences are produced not by a recursive definition of a truth predicate, but by an extended DNN language model.
From Data, to Information, to Knowledge: Measuring Knowledge Building in the Context of Collaborative Cognition
Published in Emily S. Patterson, Janet E. Miller, Macrocognition Metrics and Scenarios, 2018
Stephen M. Fiore, John Elias, Eduardo Salas, Norman W. Warner, Michael P. Letsky
So beginning with the relatively straightforward truth condition: in order for someone to know something, that something must be true. If someone claims to know something that turns out to be false, it turns out that that person only thought he knew that thing, when, in fact, he did not truly know it at all. This condition may be seen as loosely mapping onto the category of data in the DIK-T hierarchy, insofar as data is in some sense understood as pure unadorned objectivity, untouched by human processing. But what at first seemed comparatively uncontroversial now appears problematic, for how can we make sense of something being true without ourselves believing it to be true? That is, how can truth be sensibly separated from belief? Here we encounter a similar problem as that with data earlier, in that it seems nonsensical to try to think about something that exists independently of our thinking about it. The philosopher Donald Davidson, though, offers a way out of this apparent impasse, stating, simply, our ‘knowledge [is] objective in the sense that [its] truth is independent of [its] being believed to be true’ (Davidson, 2001). This amounts to the recognition that believing a thing to be true does not make it true, however difficult this recognition may sometimes be. And so our access to objectivity lies precisely in our fallibility, in the basic fact that we may be wrong, that our beliefs may be true or false. This distinction of truth from belief by means of the possibility of being mistaken carries a very clear implication for programs and procedures of knowledge construction: space must be made for error, in the working toward truth, and hence genuine knowledge. For again, whatever sense of objectivity we have lies in the acknowledgement, tacit or otherwise, of our possibly being wrong, of our beliefs possibly being false, as judged against some standard of truth understood as independent of those beliefs. Therefore any process of knowledge creation, particularly in the context of problem solving, should allow for, and even encourage, the making of mistakes, in and through some checking mechanism or feedback loop, whether of trial and error, or discussion and debate.
Doxastic logic: a new approach
Published in Journal of Applied Non-Classical Logics, 2018
MODELA model is a relational structure , where D is a non-empty set of individuals (the domain), W is a non-empty set of possible worlds, is a binary alethic accessibility relation ( is a subset of ), is a ternary doxastic accessibility relation ( is a subset of ), and v is an interpretation function. is used to define the truth conditions for sentences that begin with the alethic operators and , and is used to define the truth conditions for sentences that begin with the doxastic operators and . Informally, says that the possible world is alethically (relatively) accessible from the possible world ω, and that the possible world is doxastically accessible to the individual δ from the possible world ω, or that δ can see from ω.