one dataset · paper, tables, graph, wiki
simgrammar the grammar of simulation

locator

https://arxiv.org/abs/2011.04764

notes

Ubisoft La Forge; learned navigation for grappling hooks, jetpacks, teleporters, and double jumps on maps about ten times larger than typical RL maps, one modeled on a Ubisoft AAA game; at least 90% success.

cited by

1 records

Deep reinforcement learning policies for NPC opponents and navigation

provenance

runs in log/runs
2026-10-24
created2026-10-24-techniques-rl-designtechnique