Neural MMO is a recently released research project by OpenAI. It's a Massively Multiplayer Online game, akin to Runescape. MMOs are the game format that most closely simulate the real world.
One of the goals of the project is to get contributions from other engineers and researchers.
We’ve designed this environment with open-source expansion in mind and for the research community to build upon.
I think this is important to forecast because of a tendency in AI research to optimize for benchmarks. If NeuralMMO becomes a benchmark test for other researchers, then we'd expect to see a lot more activity in this area of multi-agent research, and one indicator of the "likelihood to benchmark status" would be if many people start expanding the software/environment
How many "meaningful" contributors to Neural MMO will there be by May 1st, 2020?
- The number of contributors will be determined by the Github contributors page.
- Meaningful will be measured as code changes totaling over 300 lines of non-boilerplate code.
- I'll evaluate non-boilerplate code by looking for code changes that were not auto-generated, that were not file renames, etc. This will be subjective, I am keen to refine this to a more objective measure.
- If there are too many contributors for me to look through all of them (over 50), I will randomly sample 10% of the contributors over 300 lines, and determine the percentage who made meaningful contributions, and multiply the total by that percentage.