Towards Goal-Oriented Agents for Evolving Problems Observed via Conversation
This work addresses the challenge of developing goal-oriented conversational agents for dynamic scenarios, but it appears incremental as it builds on existing DQN and reinforcement learning techniques.
The paper tackles the problem of training a chatbot to solve evolving problems via dialogue with a user who observes the problem, using a Deep Q-Network-based architecture and reinforcement learning. It explores training methods like curriculum learning and modified reward functions to improve performance as environment complexity increases.
The objective of this work is to train a chatbot capable of solving evolving problems through conversing with a user about a problem the chatbot cannot directly observe. The system consists of a virtual problem (in this case a simple game), a simulated user capable of answering natural language questions that can observe and perform actions on the problem, and a Deep Q-Network (DQN)-based chatbot architecture. The chatbot is trained with the goal of solving the problem through dialogue with the simulated user using reinforcement learning. The contributions of this paper are as follows: a proposed architecture to apply a conversational DQN-based agent to evolving problems, an exploration of training methods such as curriculum learning on model performance and the effect of modified reward functions in the case of increasing environment complexity.