Comonad gives us neighborhood operations for images.

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

• Comonadconcept0.785
  Dual to Monad; provides neighborhood operations (coreturn, cojoin) useful for spatial image transformations.
• Feature neighborhood exploration via cosine similarity of decoder weightsmethod0.741
  Identifying related features by cosine distance in SAE decoder space.
• If every neighborhood maintained an ongoing, updated diagnosis on a computer, all future acts and capital expenditure could be continuously guided toward improvement of bad spots and enhancement of better spotshypothesis0.739
  Alexander's proposal for institutionalizing the diagnosis-feedback-repair loop at city scale.
• The pedestrian realm is the framework for the growing neighborhood.claim0.738
  Establishes the yellow structure as the primary spatial skeleton.
• Ongoing Computerized Neighborhood Diagnosisconcept0.733
  Alexander's proposal that every neighborhood maintain an updated computer-based diagnosis to guide all future capital expenditure.
• All six of the illustrated projects help to form hulls of public space by strengthening their centeredness.claim0.721
  Claim that each example contributes to the spatial hulls described in chapter 3.
• Modern urban development depends on images to attract investors and buyers, making the financial core consistent with image-driven ideology and adverse to organic harmony.claim0.717
  Financial mechanism of image-driven development.
• When living processes are applied systematically to the public space in a human community, they will generate a system of articulated, useful, coherent, and mainly pedestrian spaces.claim0.714
  Core claim about the generative power of living process on public space.