Difference between revisions of "Neural aesthetics"

From Monoskop
Jump to navigation Jump to search
Line 55: Line 55:
 
; Introductions
 
; Introductions
 
* Kyle McDonald, [https://medium.com/@kcimc/a-return-to-machine-learning-2de3728558eb "A Return to Machine Learning"], ''Medium'', 7 Oct 2016.
 
* Kyle McDonald, [https://medium.com/@kcimc/a-return-to-machine-learning-2de3728558eb "A Return to Machine Learning"], ''Medium'', 7 Oct 2016.
 +
; Research papers
 +
* Ahmed Elgammal, et al., [https://arxiv.org/abs/1706.07068 "CAN: Creative Adversarial Networks, Generating 'Art' by Learning About Styles and Deviating from Style Norms"], 2017.
 
; Resources
 
; Resources
 
* [http://aiexperiments.withgoogle.com A.I. Experiments], a resource on machine learning art developed by Alexander Chen, creative director at Google Creative Lab, with Yotam Mann, C Christiansen, Jonas Jongejan, Gene Kogan, Kyle McDonald, et al. [https://twitter.com/alexanderchen/status/798677118473281537 Launched] on 16 Nov 2016.
 
* [http://aiexperiments.withgoogle.com A.I. Experiments], a resource on machine learning art developed by Alexander Chen, creative director at Google Creative Lab, with Yotam Mann, C Christiansen, Jonas Jongejan, Gene Kogan, Kyle McDonald, et al. [https://twitter.com/alexanderchen/status/798677118473281537 Launched] on 16 Nov 2016.

Revision as of 22:49, 27 November 2018

Partial results of the "hello world" test of a machine learning algorithm, automated recognition of handwritten digits, showing 125 test cases that the network got wrong. Each case is labeled by the network’s guess. The true classes are arranged in standard scan order. Source: Hinton et al 2006.
Basic structure of a neural network. Several techno-logical forms can be identified in the concept: scansion, that is discretisation or digitisation since the age of radio, TV, etc.; feedback loop, or the basic concept of cybernetics; and network form, here inspired by biological neurons. Diagram by Matteo Pasquinelli with Lukas Rehm, 2017. Source.

A resource on recent work between art/design and artificial neural networks in machine learning.

Events

Artists, designers

Initiatives

Literature, data, resources

Courses and textbooks for artists

Textbooks
Video lectures
  • Gene Kogan, "Machine Learning for Artists", course taught at ITP, New York University, New York, Spring 2016. 6 lectures, ~15 hours.
  • Gene Kogan, "The Neural Aesthetic", course taught at School Of Machines, Making, and Make-Believe, Berlin, Summer 2016. 8 lectures, ~18 hours.
Introductions
Research papers
Resources
  • A.I. Experiments, a resource on machine learning art developed by Alexander Chen, creative director at Google Creative Lab, with Yotam Mann, C Christiansen, Jonas Jongejan, Gene Kogan, Kyle McDonald, et al. Launched on 16 Nov 2016.
  • People + AI Research, Google Design, 2018ff.

Data sets

Scientific introduction into deep learning

Historization of deep learning

Recent debates on artificial intelligence in the humanities and social sciences

Art practice and criticism