The tendency of an artificial neural network to forget previously learned information when acquiring new information is called…
Catastrophic forgetting, also known as catastrophic interference, is a common problem in many artificial neural network models. It refers to the catastrophic loss of previously learned responses when a network is trained with a single new response. This is a primary reason why artificial neural networks are not able to continuously learn from their surroundings and must be fully trained on an expected set of responses before being put into service.
Although catastrophic forgetting is a specific type of interference, interference is a more general phenomenon observed in natural learning systems such as humans, which tends to cause gradual losses. In contrast, losses caused by interference in artificial systems are catastrophic.
The stability-plasticity problem is sometimes promoted as a dilemma to explain away the problem of catastrophic forgetting in traditional artificial neural networks, even though natural systems have overcome it. More recently, the problem has been fully overcome in artificial neural networks through multi-temporal synapses.
Sol Rezza’s work, Catastrophic Forgetting, addresses this concept, raising questions about the implications of machine memories in a posthuman future.
In a world where machines can remember infinitely while humans no longer exist, What good will all that information do? Will the machines be nostalgic for memories that never really lived?
While the problem of catastrophic forgetting has been overcome, the potential impact of machine memories on the sociotechnological future remains a matter of debate.
Catastrophic Forgetting is an audiovisual installation that was designed as a site-specific work to be experienced live with 25 audio channels. The project debuted on November 21, 2021, as part of the unveiling of the new multi-channel sound system that was permanently installed at the X-Church cultural center.
This unique project is installed in the 1882 Gothic church, where a 25-channel discrete audio system has been installed. The audio system is one of the largest multi-channel systems in Lincolnshire, UK. This gives local artists and X-Church residents the opportunity to create their own spatial audio works using this state-of-the-art system.
For the spatial audio configuration, Sol Rezza used the SoundSquares plugin designed by Daz Disley.
Premiere: 27.11.2021 X-church, Gainsborough, UK. | Cold Diffusion Festival
Year of production: 2021
Running time in mm:ss: 17:42
Video Format: Digitial | Full HD | .MOV H.264 .MP4
Audio Video: Binaural
Audio Performance: Multichanel Performance 25.1
Audio Plugin: SoundSquares
Idea and realization: Sol Rezza
Audio Engineer and Supervisor: Daz Disley
Technical Support: Slumgothic/X- Church Crew | Irina Vainio
Voices: Lena Sass Hughes
Cold Diffusion Artist: Daz Disley, Sol Rezza, Franco Falistoco, Duncan Chapman, Emily Welther
Photo and video: Fenia Kotsopoulou | Irina Vainio