Interpreting Tangled Program Graphs Under Partially Observable Dota 2 Invoker Tasks.
Robert J. SmithMalcolm I. HeywoodPublished in: IEEE Trans. Artif. Intell. (2024)
Keyphrases
- reward function
- partially observable
- markov decision processes
- reinforcement learning
- state space
- partial observability
- markov decision problems
- partially observable domains
- partial observations
- partially observable environments
- action models
- infinite horizon
- special case
- transfer learning
- belief space
- linear programming
- multi agent