A Machine with Short-Term, Episodic, and Semantic Memory Systems
This work addresses memory modeling in AI agents, but it is incremental as it applies existing deep Q-learning to a new environment with a structured memory approach.
The paper tackled the problem of designing an agent with human-like memory systems (short-term, episodic, and semantic) using knowledge graphs, and showed that this agent outperforms one without such memory in a custom reinforcement learning environment called 'the Room'.
Inspired by the cognitive science theory of the explicit human memory systems, we have modeled an agent with short-term, episodic, and semantic memory systems, each of which is modeled with a knowledge graph. To evaluate this system and analyze the behavior of this agent, we designed and released our own reinforcement learning agent environment, "the Room", where an agent has to learn how to encode, store, and retrieve memories to maximize its return by answering questions. We show that our deep Q-learning based agent successfully learns whether a short-term memory should be forgotten, or rather be stored in the episodic or semantic memory systems. Our experiments indicate that an agent with human-like memory systems can outperform an agent without this memory structure in the environment.