What Is the Sublinear Computation Paradigm?

This chapter introduces the “sublinear computation paradigm.” A sublinear-time algorithm is an algorithm that runs in time sublinear in the size of the instance (input data). In other words, the running time is o(n), where n is the size of the instance. This century marks the start of the era of big data. In order to manage big data, polynomial-time algorithms, which are considered to be efficient, may sometimes be inadequate because they may require too much time or computational resources. In such cases, sublinear-time algorithms are expected to work well. We call this idea the “sublinear computation paradigm.” A research project named “Foundations on Innovative Algorithms for Big Data (ABD),” in which this paradigm is the central concept, was started under the CREST program of the Japan Science and Technology Agency (JST) in October 2014 and concluded in September 2021. This book mainly introduces the results of this project.