Reinforcement learning methods have been used for learning dialogue policies from the experience of conversations. However, learning an effective dialogue policy frequently requires prohibitively many conversations. This is partly because of the sparse rewards in dialogues, and the relatively small number of successful dialogues in early learning phase. Hindsight experience replay (HER) enables an agent to learn from failure, but the vanilla HER is inapplicable to dialogue domains due to dialogue goals being implicit (c.f., explicit goals in manipulation tasks). In this work, we develop two complex HER methods providing different trade-offs between complexity and performance. Experiments were conducted using a realistic user simulator. Results suggest that our HER methods perform better than standard and prioritized experience replay methods (as applied to deep Q-networks) in learning rate, and that our two complex HER methods can be combined to produce the best performance.
Machine-learning based dialogue managers are able to learn complex behaviors in order to complete a task, but it is not straightforward to extend their capabilities to new domains. We investigate different policies' ability to handle uncooperative user behavior, and how well expertise in completing one task (such as restaurant reservations) can be reapplied when learning a new one (e.g. booking a hotel). We introduce the Recurrent Embedding Dialogue Policy (REDP), which embeds system actions and dialogue states in the same vector space. REDP contains a memory component and attention mechanism based on a modified Neural Turing Machine, and significantly outperforms a baseline LSTM classifier on this task. We also show that both our architecture and baseline solve the bAbI dialogue task, achieving 100% test accuracy.
Training a task-completion dialogue agent via reinforcement learning (RL) is costly because it requires many interactions with real users. One common alternative is to use a user simulator. However, a user simulator usually lacks the language complexity of human interlocutors and the biases in its design may tend to degrade the agent. To address these issues, we present Deep Dyna-Q, which to our knowledge is the first deep RL framework that integrates planning for task-completion dialogue policy learning. We incorporate into the dialogue agent a model of the environment, referred to as the world model, to mimic real user response and generate simulated experience. During dialogue policy learning, the world model is constantly updated with real user experience to approach real user behavior, and in turn, the dialogue agent is optimized using both real experience and simulated experience. The effectiveness of our approach is demonstrated on a movie-ticket booking task in both simulated and human-in-the-loop settings.
This paper proposes a deep neural network model for joint modeling Natural Language Understanding (NLU) and Dialogue Management (DM) in goal-driven dialogue systems. There are three parts in this model. A Long Short-Term Memory (LSTM) at the bottom of the network encodes utterances in each dialogue turn into a turn embedding. Dialogue embeddings are learned by a LSTM at the middle of the network, and updated by the feeding of all turn embeddings. The top part is a forward Deep Neural Network which converts dialogue embeddings into the Q-values of different dialogue actions. The cascaded LSTMs based reinforcement learning network is jointly optimized by making use of the rewards received at each dialogue turn as the only supervision information. There is no explicit NLU and dialogue states in the network. Experimental results show that our model outperforms both traditional Markov Decision Process (MDP) model and single LSTM with Deep Q-Network on meeting room booking tasks. Visualization of dialogue embeddings illustrates that the model can learn the representation of dialogue states.
The amount of dialogue history to include in a conversational agent is often underestimated and/or set in an empirical and thus possibly naive way. This suggests that principled investigations into optimal context windows are urgently needed given that the amount of dialogue history and corresponding representations can play an important role in the overall performance of a conversational system. This paper studies the amount of history required by conversational agents for reliably predicting dialogue rewards. The task of dialogue reward prediction is chosen for investigating the effects of varying amounts of dialogue history and their impact on system performance. Experimental results using a dataset of 18K human-human dialogues report that lengthy dialogue histories of at least 10 sentences are preferred (25 sentences being the best in our experiments) over short ones, and that lengthy histories are useful for training dialogue reward predictors with strong positive correlations between target dialogue rewards and predicted ones.