The need for magic: On artificial intelligence, translation & navigation

26.01.2019 15:48
by Auris-E. Lipinski
Alt-Tag here

Date: Saturday, 26. January 2019

Time: 12:00 o`clock (ends at 18:00)

Location: Daimler Wohnzimmer

(Access via Helmholtzstrasse 2-9, opposite of Morsestraße, through the barrier (only on foot or bike) to the end of the building site, the big wooden front porch on the left)

 

 

 

Programm:

12:00 h Introduction by Alexander Mankowksy: the cultural desire to aid magical ai

13:00 h Presentation by Auris-E Lipinski: translation as (magical) transformation in navigation and path descriptions

14:00 h Discussion

14:30 h Pause

15:00 h Presentation by Katja Stepec: translation as a criterion for artificial intelligence.

16:00 h Discussion

16:30 h Pause

16:40 Discussion of all presentations

Moderation by Thomas Sprengler

 

 

 

 

Translation as (magical) transformation in navigation and path descriptions

Speaker: Auris Lipinski

Translation can be understood as transformation with regard to navigation, in difeerent colours and kinds. This observation sheds some light on nouanced differences between practical and theoretical endeavours in path finding descriptions and how meaning is here dependend on correct or meaningful translation. It leads, along the lines of the example of navigation, to an explanation which puts emphasis on the importance of embodiment and enactivism in meaning construction. It ends with a take on what preconditions artificial systems would have to have available, to be even remotely able to approximate anything that leads to (translation) success in path finding descriptions.

 

Translation as a criterion for Artificial Intelligence.

I want to defend the thesis, that an AI can be defined sufficiently by its ability to translate. Another way to put it: A computer is an AI, if it is able to translate. To defend this argument, one has to answer two questions:

a. Under what conditions are non-human devices able to translate between natural or general languages? A first answer is given with reference to an israeli philospher and linguist, Bar-Hillel, and with a survey of modern approaches of Machine Translation (MT). Bar-Hillel proposed in the 1950s that MT won´t be succesfull and I think his suggestion is still valid, because of reasons concerning the explanation of meaning and language in a philosophy of language.

b. What exactly is understood by translation? I will suggest that MT is only successful when translation is understood as a correlation, while human translation has to be understood as a kind of practice which is essentially normative. This argument is based on the philosophical approaches of W. v. O. Quine and Robert Brandom. Presupposing, that an AI is to be understood not to be just MT but more like a human translator, a sufficient definition for AI therefore includes the ability to translate in form of a normative practice. This proposition also includes two different approaches to explain meaning philosophically. While translation as correlation is based on an atomistic approach to language, translation as a normative practice is based on a holistic approach to language. I understand general or natural languages to be holistic and therefore in need of a translation as a normative practice.

 

Speaker: Katja Stepec

 

Video, pictures- and sound recordings:
Additional to the audio recordings, we started to film our event content, because it is precious. The materials of the workshops are kept and made available, so far, for free. Please do tell us, if there are comments on the format.

***If you attend, you agree to be on these recordings. ***


Data use permission:
When you come to the event and/ or leave your email – or other messenger-contact – with us, you have done this with the permission to invite you again. Please let us know immediately, if this is invitation should not have reached you in the way it did. We need written consent, so if you answer to this email with a simple “Yes, you may invite me”, we are good. If you do not do so, we will at some point come to you personally and ask, if we can. If we cannot, you will not receive further invitations.

Undo data use permission:
If you have received an e-mail but want to undo your data permission, or have any trouble with the way it is realised, please let us know. You will not get an invitation again, and we will try to cater to your needs as much as possible, if you have comments on details that would make the conduct more agreeable to you.

Go back

© 2015 PhenCoCo| Contact | Impressum

Aktuelle Themen

Hier finden Sie einige HOT-Topics aus dem VIOM Protfolio

Optimierung

Online-Demo

Sehen Sie selbst

Optimierung2

Online-Demo

Sehen Sie selbst

Optimierung3

Online-Demo

Sehen Sie selbst