Black-Box Nature of Deep Learning

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Deep Learningconcept0.756
  Learning hierarchical representations of non-decomposable functions; proposed as formal equivalent to ETI process.
• Deep Reinforcement Learningmethod0.732
  AI training method inspired by behaviorism, used for autonomous cars and drones; cited as bioinspired success
• Is deep learning at a similar, albeit more modest, transition point as the invention of the microscope?question0.725
  Motivating analogy question framing the circuits agenda as a potential paradigm shift in interpretability
• Unsupervised learning builds a low-dimensional model of the input data.claim0.725
  Clarifies what unsupervised learning does.
• Human consciousness is tied to a biological learning algorithm — the simplest algorithm discoverable by evolutionary search to train a self-organizing biological substrate to become intelligent.claim0.723
  Core theoretical claim connecting consciousness to biological learning
• We hypothesize that consciousness may be at the heart of a universal biological learning algorithm — one that runs on self-organizing groups of communicating cells sharing evolutionary incentives — and that it creates rather than results from organized mental architecture.hypothesis0.722
  The Genesis Hypothesis as explicit predictive conjecture
• TAME does not claim that mind is inevitably baked in regardless of physical implementation; causal structure and cybernetic properties are key determinants.claim0.722
  Distinguishes TAME from panpsychist views; emphasizes role of organization.
• Deep networks are biased toward finding simple fits to the data, and the bigger the model the stronger the bias, driving convergence to a smaller solution spacehypothesis0.722
  Selective pressure toward convergence via implicit regularization