hexerei - the art or power of bringing magical or preternatural power to bear or the act or practice of attempting to do so

• make a blog visual test page that shows labeling and various tests
  

geojson.io - mapbox open source geojson editor geojson.xyz - sample files via natural earth geonmaes.org - collection of mostly points for many features geoman.io - alternative freemium geojson editor

geojson => hex map

specials:

Geometry Line with id='origin' with hex side length = pointy radius, and direction 90deg right of 'up' orientation, or Point with properties 'scale' hex size in km, and orient=degrees default 0 default flat top, optional pointy=true.

Polygon (or multipolygon) geometry with id='boundary'

other features optional property map='tile|edge|vertex|none', default tile optional property label=str

coordinate systems:

[web mercator][webgm]for geojson e.g. for map background tiles or cities [webgm]: https://en.wikipedia.org/wiki/Web_Mercator_projection

cube/axial coordinates as native hex coords

user coords (string ids) like offset xxyy "0304" or alternative axial like "AA01" needs a cube2id and id2cube method

map geojson objects to hex tiles, edges and/or vertices via shapely

build and run like

npx tsc node --es-module-specifier-resolution=node dist/flowsnake.js

• Hold the Line
• Drive on Metz
• Target Arnhem
• Gettysburg 150
• Battle for Moscow

• [AI and wargaming][https://arxiv.org/pdf/2009.08922.pdf]

• Playing the Game of Risk with an AlphaZero Agent

• From Poincare Recurrence to Convergence in Imperfect Information Games: ´ Finding Equilibrium via Regularization

• Mastering the Game of Stratego with Model-Free Multiagent Reinforcement Learning

• Snake reinforcement learning

• https://github.com/baskuit/R-NaD

• NeuroHex: A Deep Q-learning Hex Agent [slides]](https://webdocs.cs.ualberta.ca/~hayward/talks/hex.deepQ.pdf), paper

• AI in Human-computer Gaming: Techniques, Challenges and Opportunities

• file:///Users/psurry/Downloads/algorithms-15-00282.pdf

• GameAIPro

• Neural Replicator Dynamics: Multiagent Learning via Hedging Policy Gradients

Uber's H3 https://h3geo.org/ use directed edge identifiers between two hexes to represent flow between them?