research thread:

groworld_hpi > Boskoi 2010 > Augmented Ecology 2014 > Machine Wilderness 2015 > Environmental Machine Learning (2018)

project website:

potential starting questions:

  • if/how the concept of the 'umwelt' in biological creatures relates to the 'world view' that forms in artificial neural networks during training.
  • how do animals, plants or machines learn through experience and exposure? (+cognitive biases)
  • (how) could an AI become environmentally literate? (+ implications)
  • what does a 'synthetic worldview' mean for the understanding/appreciation of environmental complexity?
  • how do strategies of environmental observation compare/relate (in AI, choreography, ecology, art, landscaping, traditional cultures,..)
  • who is the observer in these experiments? what kind of power-relations come out? (+symbiogenesis)
  • thalience: can environments be given their own voice? (or one that can reach modern humans on more level power-relations)

blurb:

Complex machines have been part of our environment for many centuries. Pioneers like al Jazari already made programmable automata around 1200AD. Machines came to dominate the land, sea and air dramatically since the Industrial Revolution. Until very recently the ability to relate to the environment was limited to plants and animals, but now machines are starting to blur those lines. What does it mean if machines join animals and plants there on more equal levels of awareness? Environmental Machine Learning is a program of fieldwork sessions with experiments as vehicles for materialising questions.

context:

All mayor tech companies have made AI their top priority, some say in a race to file patent applications. In any case these systems are not just reaching into the depths of human society and media, but also root deeply into the remotest mangroves, deserts or reefs. Some first experiments with machine learning have been undertaken by ecologists. EML aims for a fundamental exploration of environmental literacy and how this could be made accessible to / obtained by an AI.

methods:

  • fieldwork: in-situ exploration through interactions between man-machine-environment
  • prototyping: like in Boskoi & Machine Wilderness with experiments as vehicles for materializing questions
  • critical reflection
  • multimodal and transdisciplinary approach: could the project also give room to explore observation strategies from various domains of human inquiry and probe them in-situ?

program: (under construction)

  • dec 2017 / nov 2018 > EML Meetup series at MidWest Experimental Station Amsterdam > theme: Synthetic Environmental Literacy
  • mar 2018 > fieldwork session Terschelling > theme: Random Forests: Environmental observation and perception into algorithm > 8 ppl / 3 or 4 days
  • may 2018 > fieldwork session Finland > theme: Rules of Engagement: Machine and Animal interactions > 4 ppl / 10 days
  • nov 2018 > critical reflection / writing, web or print > Fieldguide to Environmental Machine Learning
  • exhibition (Artis Zoo?)
  • Plain Air Nouveau EU program

reading:

Youtube Lectures:

framing:

initiatives:

code of conduct:

Algorithms underpin the global technological infrastructure that extracts and develops natural resources such as minerals, food, fossil fuels and living marine resources. They facilitate global trade flows with commodities and they form the basis of environmental monitoring technologies. Last but not least, algorithms embedded in devices and services affect our behavior - what we buy and consume and how we travel, with indirect but potentially strong effects on the biosphere. As a result, algorithms deserve more scrutiny.
- Victor Galaz / Stockholm Resilience Centre

see also:

  • environmental_machine_learning.txt
  • Last modified: 2020-07-03 10:44
  • by theunkarelse