Reinforcement Learning in POMDPs with Memoryless Options and Option-Observation Initiation Sets
This addresses the problem of learning in partially observable hierarchical environments for reinforcement learning practitioners, offering a simpler and more efficient approach compared to existing methods.
The paper tackled the combined problem of hierarchy and partial observability in reinforcement learning by introducing Option-Observation Initiation Sets (OOIs), which condition option initiation on the previous option, showing they are at least as expressive as Finite State Controllers and enabling optimal policies in challenging POMDPs with improved sample efficiency over recurrent neural networks.
Many real-world reinforcement learning problems have a hierarchical nature, and often exhibit some degree of partial observability. While hierarchy and partial observability are usually tackled separately (for instance by combining recurrent neural networks and options), we show that addressing both problems simultaneously is simpler and more efficient in many cases. More specifically, we make the initiation set of options conditional on the previously-executed option, and show that options with such Option-Observation Initiation Sets (OOIs) are at least as expressive as Finite State Controllers (FSCs), a state-of-the-art approach for learning in POMDPs. OOIs are easy to design based on an intuitive description of the task, lead to explainable policies and keep the top-level and option policies memoryless. Our experiments show that OOIs allow agents to learn optimal policies in challenging POMDPs, while being much more sample-efficient than a recurrent neural network over options.