Keywords: Adversarial images, machine learning, deep neural networks, artificial intelligence, invisibility, perception, operative image, algorithms, digital cryptography, decipherability
Adversarial images, inputs designed to produce errors in machine learning systems, are a common way for researchers to test the ability of algorithms to perform tasks such as image classification. “Fooling images” are a common kind of adversarial image, causing mis-categorisation errors which can then be used to diagnose problems within an image classification algorithm. Situations where human and computer categorise an image differently, which arise from adversarial images, reveal discrepancies between human image interpretation and that of computers. In this paper, aspects of state of the art machine learning research and relevant artistic projects touching on adversarial image approaches will be contextualised in reference to current theories. Harun Farocki’s concept of the operative image will be used as a model for understanding the coded and procedural nature of automated image interpretation. Through comparison of current adversarial image methodologies, this paper will consider what this kind of image production reveals about the differences between human and computer visual interpretation.
- Rosemary Lee is an artist and researcher whose work investigates interrelations between technologies and processes of natural science. Their work brings together influences from media geology, hybrid ecology and posthumanism through theory-driven practice-led research. A selection of their notable exhibitions include machines will watch us die (Holden Gallery, GB, 2018), A New We (Kunsthall Trondheim, NO, 2017), Hybrid Matters (Nikolaj Kunsthal, DK, 2016), TRANSART (Dome of Visions, DK, 2015) and Artifacts (Palais des Beaux Arts, AT, 2015). Rosemary Lee is currently a PhD fellow at the IT University of Copenhagen. She has also acted as artist-researcher in residence in international contexts including the Ayatana Arist Research Residency (CA, 2017), rural.scapes – Laboratory in Residence (BR, 2016) and the transmediale Vilém Flusser Archive Residency for Artistic Research (DE, 2014).
Full text p. 321 – 324