AAAI 2013 Fall Symposium Proposal

Title – Self, Intelligent Reflection & Implementing Safe Autopoiesis

Description – In I Am a Strange Loop, Douglas Hofstadter argues that the key to understanding selves is the “strange loop”, a complex feedback loop inhabiting our brains and, arguably, constituting our minds. Artificial Intelligence (AI) and Artificial General Intelligence (AGI) generally have a focus of solving problems or achieving goals.  Instead, this implementation-oriented symposium will focus on this loop and guided self-(re)creation – particularly as a method of achieving human-level intelligence through iterative improvement (“seed AI”).  Ideally, the ultimate end product is a successful AGI with extensive self-knowledge and a safe and/or ethical motivational system promoting cooperation with and contribution to community via improvement of self and tools.

The AAAI communities that are expected to find this appealing are sub-groups of the Artificial General Intelligence (AGI) community, the Biologically Inspired Cognitive Architectures (BICA) community, and the Machine Consciousness (MC) community.  Non-AAAI communities that are expected to find this appealing are the NeuroBioEthics  community and the Ethical/Moral Machines sub-group of the Philosophy & Technology community.

This symposium is proposed as an implementation-orientation exploration of self, to include:

  • integrative architectures that are explicitly goal driven
    • “self” as operating system
    • “self” as society (of mind – Marvin Minsky)
    • “self” as economy (of idiots – Eric Baum)
    • “self” as authorship (Dennett, Wegner)
  • “safe” and/or ethical motivational systems
    • goal hierarchies
    • “safe” goal content
    • evaluation schemes
  • reflection
    • self-examination
    • self-modeling & self-knowledge
    • goal-based self-evaluation for self-improvement
  • emotions
    • as knowledge/rules of thumb/”actionable qualia”
    • as helpful & unhelpful biases (and how to intelligently improve)
    • as evaluation & enforcement mechanisms
  • integrating different knowledge and action representation schemes
    • coordination & translation between various schemes
    • analyzing trade-offs & knowing when to switch between schemes
  • self-improvement
    • via automated tool/method incorporation and evaluation
    • via learning (knowledge incorporation)
      • (re-)building schemes and models
    • discovery (refactoring, modularization, encapsulation and scale-invariance)
      • Meta-Optimizing Semantic Evolutionary Search (MOSES)
      • Frequent & Interesting Sub-HyperGRAph Mining (FISHGRAM)

While solutions need to be grounded and extensible, the symposium would prefer approaches starting with some initial structure rather than a tabula rasa with the lowest level bootstrapping approaches or first causes explanations (except where they are fully extended to initial structures and/or used to justify such structures).  Also, while autopoiesis and “functional consciousness” are obviously key topics, we would prefer that phenomenal consciousness arguments be considered off-topic.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: