Neural Information Retrieval: A Literature Review
This is a literature review that synthesizes existing research for the IR community, identifying obstacles and future directions, but it is incremental as it does not present new experimental results.
The paper surveys the emerging field of Neural Information Retrieval (IR), highlighting the use of neural embeddings for queries and documents, but notes that deep neural networks have not yet achieved the same level of success in IR as in other areas like speech recognition or computer vision.
A recent "third wave" of Neural Network (NN) approaches now delivers state-of-the-art performance in many machine learning tasks, spanning speech recognition, computer vision, and natural language processing. Because these modern NNs often comprise multiple interconnected layers, this new NN research is often referred to as deep learning. Stemming from this tide of NN work, a number of researchers have recently begun to investigate NN approaches to Information Retrieval (IR). While deep NNs have yet to achieve the same level of success in IR as seen in other areas, the recent surge of interest and work in NNs for IR suggest that this state of affairs may be quickly changing. In this work, we survey the current landscape of Neural IR research, paying special attention to the use of learned representations of queries and documents (i.e., neural embeddings). We highlight the successes of neural IR thus far, catalog obstacles to its wider adoption, and suggest potentially promising directions for future research.