Listwise Learning to Rank Based on Approximate Rank Indicators
Keywords:Machine Learning (ML)
AbstractWe study here a way to approximate information retrieval metrics through a softmax-based approximation of the rank indicator function. Indeed, this latter function is a key component in the design of information retrieval metrics, as well as in the design of the ranking and sorting functions. Obtaining a good approximation for it thus opens the door to differentiable approximations of many evaluation measures that can in turn be used in neural end-to-end approaches. We first prove theoretically that the approximations proposed are of good quality, prior to validate them experimentally on both learning to rank and text-based information retrieval tasks.
How to Cite
Thonet, T., Cinar, Y. G., Gaussier, E., Li, M., & Renders, J.-M. (2022). Listwise Learning to Rank Based on Approximate Rank Indicators. Proceedings of the AAAI Conference on Artificial Intelligence, 36(8), 8494-8502. https://doi.org/10.1609/aaai.v36i8.20826
AAAI Technical Track on Machine Learning III