Gravitational waves are characterised by non-linear dynamics, high dimensional source systems, and present noisy signal in the ground-based detectors of LIGO and Virgo. Machine learning has emerged as a powerful approach for solving problems in astrophysics presenting such challenges. In this talk, I review applications of machine learning techniques for the analysis of gravitational-waves data, including fast measurements of the astrophysical parameters of gravitational-wave sources, algorithms for reduction and characterisation of non-astrophysical detector noise, and applications to gravitational waveforms modelling. These applications demonstrate how machine learning techniques may be harnessed to enhance the science that is possible with current and future gravitational-wave detectors.