Open domain Conversational Model using transfer learning

Conversational modeling is a complex task, comprising of natural language understanding and generating text as a response to the input given. In recent years, there has been the rise of a new set of conversational AI systems that are based on deep-learning neural networks architectures. With the advent of bigger and larger Transformer based Language Models trained on gigabytes of textual data and billions of parameters, the language generators are now able to generate answers, write articles and summarize texts with human-like fluency. In this paper, an open domain conversational agent that leverages transfer-learning based on OpenAI’s GPT-2 transformer language model has been implemented and then fine-tunes it on the Reddit dataset to generate relevant and in-context responses. Experimental results and live talking with the system show that the sentence formation for the responses are quite natural and efficiently uses the history of chat to select the best response.