We use reinforcement learning to learn dialogue policies in a collaborative furniture layout negotiation task. We employ a variety of methodologies (i.e., learning against a simulated user versus co-learning) and algorithms. Our policies achieve the best solution or a good solution to this problem for a variety of settings and initial conditions, including in the presence of noise (e.g., due to speech recognition or natural language understanding errors). Also, our policies perform well even in situations not observed during training. Policies trained against a simulated user perform well while interacting with policies trained through co-learning, and vice versa. Furthermore, policies trained in a two-party setting are successfully applied to a three-party setting, and vice versa.
BEIJING – Japan and China on Monday held their first security dialogue involving senior diplomats and defense officials in nearly two years. The talks in Beijing took place as the two countries attempt to set up a maritime and aerial communication mechanism to prevent accidental clashes in and above the East China Sea, where China has been asserting its claim to the Japan-administered Senkaku Islands. The meeting was held before Chinese Premier Li Keqiang's potential first visit to Japan since taking office in 2013. He may visit next month to attend a trilateral summit involving the two countries and South Korea. Kong Xuanyou, China's assistant foreign minister, said he hopes the dialogue will play an "active role in enhancing the momentum of improving ties between the two countries."
Chatbots can help you order pizza, accept payments and be super racist, but their usefulness has been pretty limited. However, Facebook announced today that it has created a much more capable bot by giving it the ability to negotiate, strategize, and plan ahead in a conversation. Getting computers to understand conversation at a human level has been a pretty unsuccessful venture thus far. It requires not only a large amount of knowledge but rapid and accurate adaptability as well. But researchers at Facebook Artificial Intelligence Research (FAIR) have developed a new technique that lets bots successfully navigate a very human type of dialogue -- negotiations.
Dialogue management is the process of deciding what to do next in a dialogue. In human communication, dialogue management is a process that seems to go unnoticed much of the time; only rarely are we away of having to make a decision on what to say next, or on whether to say anything at all. When designing a computer dialogue system, however, we either have to pre-program the possible dialogues according to certain fixed sequence of utterances, or else we have to define a'dialogue manager', who decides what the system should do next based on a model of the current dialogue context. Research on the design of intelligent computer dialogue management systems has also inspired investigations of human dialogue management. These investigations have made it clear that human dialogue management is actually highly complex and sophisticated, and suggests that one of the stumbling blocks for the development of high-quality speech dialogue systems is the design of dialogue managers that have some of the sophistication and subtlety of human dialogue management. In this paper, we will examine the notion of'dialogue context' in the sense of the information that is relevant for deciding what to do next in a dialogue.
An important difficulty in developing spoken dialogue systems for robots is the open-ended nature of most interactions. Robotic agents must typically operate in complex, continuously changing environments which are difficult to model and do not provide any clear, predefined goal. Directly capturing this complexity in a single, large dialogue policy is thus inadequate. This paper presents a new approach which tackles the complexity of open-ended interactions by breaking it into a set of small, independent policies, which can be activated and deactivated at runtime by a dedicated mechanism. The approach is currently being implemented in a spoken dialogue system for autonomous robots.