ajayn1997/Self-Critical-Sequential-training-with-RL-for-chatbots

A chatbot implemented as a seq-2-seq model and trained using cross entropy method. The performance of the chatbot is improved by using Sequence Level Training using REINFORCE algorithm. In order to apply the REINFORCE algorithm (Williams, 1992; Zaremba & Sutskever, 2015) to the problem of sequence generation we cast our problem in the reinforcement learning (RL) framework (Sutton & Barto, 1988). Our generative model (the RNN) can be viewed as an agent, which interacts with the external environment (the words and the context vector it sees as input at every time step). The parameters of this agent defines a policy, whose execution results in the agent picking an action. In the sequence generation setting, an action refers to predicting the next word in the sequence at each time step. After taking an action the agent updates its internal state (the hidden units of RNN). Once the agent has reached the end of a sequence, it observes a reward. We can choose any reward function. Here, we use BLEU (Papineni et al., 2002) and ROUGE-2 (Lin & Hovy, 2003) since these are the metrics we use at test time.