I just arrived at the Centre d'Art i Natura in Catalonia, where I will stay from June 4th-18th to work on a fifth and last intervention as part of Absences. I have to say I didn't expect the place to be so gorgeous. It is also perfecly calm and isolated, which will allow me to work on this last project. But will this really be the last one? I find this concept problematic: how can I plan that this will be the last of the series? Why?
Sure, there is still a lot of explorations to do with such artificial-natural "things". For instance, I'm not even beginning to solve some of the basic issues such as the energy management. For what I know, none of the four interventions that I have made thus far have survived for long in their "natural habitat". And I am fine with it, as this ephemeral character is also part of the project.
But I want to do more. I want to spend more time experimenting. I want to understand better what is going on. I want to do more research on energy management and adaptivity. I want to work on concepts such as autopoiesis and performativity, I wanna look at the phenomenological perspective of those "artificial life forms". I don't wanna talk about solitude and separation: I want to articulate those artificial art forms around a post-humanist vision of the contemporary world. Oh my ... ok, more practically, I feel like on one side, low-fi devices are interesting for experiments but aren't robust to the outdoors. They're hard to monitor, which makes it hard to accumulate data and see what is going on, especially when I have to leave at the end of a residency. I want to spend more time monitoring, looking at the long term effects in less hostile environments. Then perhaps, later on, I can concentrate on the material approaches that can allow me to build devices that can stand the test of time. My new collective work Vessels is an example of a good compromise between low-fi and robustness (the devices are left outside only 4-6 hours at a time).
But anyway. I'm here now. I wanted to do one last project as part of this series because, despite all I have just said, I really like the project. This time, I have a very tight schedule, but I also already a pretty clear idea of what I wanna do. I'm going to work on a simple robotic system that can basically control the orientation of a set of solar panels. It will try to optimize its energy input by "following the sun" all through the day. However, there will be a "counter-objective" which I have yet to define. One idea is that, after looking at the sun for a while, it will start "shaking" (thus spending energy) and will only be able to stop the shaking by "looking away" from the sun.
Very importantly, the agent will have no a priori knowledge: it will learn how to behave in the most optimal way by using a Machine Learning algorithm known as Reinforcement Learning (RL). Basically, this kind of system first takes random actions but is able to learn what are the best sequences of action by adapting to rewards and punishments it gets.
I took a walk around tonight, the Pyrenees in the sunset light were so beautiful, it was almost cliché. There even was a rainbow (seriously). I almost missed dinner with Lluís (the director of the center) and Océane (the other artist in residence).I just arrived at the Centre d'Art i Natura in Catalonia, where I will stay from June 4th-18th to work on a fifth and last intervention as part of Absences. I have to say I didn't expect the place to be so gorgeous. It is also perfecly calm and isolated, which will allow me to work on this last project. But will this really be the last one? I find this concept problematic: how can I plan that this will be the last of the series? Why?