“Xepa: Autonomous Intelligent Light And Sound Sculptures That Improvise Group Performances” presented by Galanter


Presentation Title:

  • Xepa: Autonomous Intelligent Light And Sound Sculptures That Improvise Group Performances



  • XEPA is the name of both the art project and individual intelligent sculptures that display animated colored light sequences and produce music and sound. [1] XEPA is an acronym for “XEPA Emerging Performance Artist”; the performance is emergent; the “young” XEPAs are emerging artists.
    Starting in a unique random state, each XEPA “watches” the others (via data radio) and modifies its own aesthetic behavior to create a collaborative improvisational performance. In doing so each XEPA independently evaluates the aesthetics of the other sculptures, infers a theme or mood being attempted, and then modifies its own aesthetics to better reinforce that theme. Each performance is unique, and a wide variety of themes and moods can be explored.
    It’s important to note that while data radio is used, it is in principle the same as each XEPA watching the behavior of the others. The radio messages sent are merely descriptions of what that particular XEPA is doing at that particular time. No actions occur by command, and there is no script to follow.
    While XEPA is fundamentally an artwork, it provides a robust platform for artificial intelligence experiments in computational aesthetic evaluation (CAE). [2] The previously shown XEPA 1.0 was purely visual. XEPA 2.0 proposed for ISEA 2015 adds sound for the first time, greatly expanding the expressive gamut and complexity.

    Thematic Statement
    Many writers such as Boden emphasize that novelty is a necessary but insufficient criteria for creativity. [3] Creativity also carries with it the implication that the results are useful or otherwise of value. To fully qualify as being creative, computers will have to apply a critical function that makes decisions as to which options are of greater or lesser value.
    XEPA anticipates a future where creative machines form their own societies. Going beyond mere generative art, future systems will exhibit artistic creativity with the addition of artistic judgment via computational aesthetic evaluation. And rather than using low level literal protocols, these systems will communicate at higher levels of abstraction by observing each other’s behavior. Just as humans communicate using high-­‐level semantics in language and art, creative digital systems will do the same.
    XEPA is offered as the next step forward in generative art, with the addition of a critical function to the somewhat mindless autonomy of genetic algorithms, L ­‐- systems, cellular automata, and so on. It shifts the emphasis from the old text of low­‐- level code to the new text high-­‐level semantics and behavior. It presents an artwork as prototype; as a DIY platform for learned behaviors of increasing complexity.


    1. Galanter, P., Xepa – Autonomous Intelligent Light And Sound Sculptures That Improvise Group Performances. Leonardo, 2014. 47(4): p. 386-393.
    2. Galanter, P., Computational Aesthetic Evaluation: Past and Future, in Computers and Creativity, J. McCormack and M. dinverno, Editors. 2012, Springer: Berlin.
    3. Boden, M.A., The creative mind: myths and mechanisms. 2nd ed. 2004, London; New York: Routledge. xiii, p. 344

Project Website: