A summary of existing work on the use of machine learning and deep learning methods in biometrics is presented here. Biometrics traits covered include physiological (image, voice) as well as behavioral (gait, signature) features. This study shows that machine learning has a high potential to improve the performance of biometrics systems due to ML’s ability to mine, search and analyze big sets of data, performing matching tasks more quickly and reliably than the conventional methods. At the end some key challenges in use of adoptingML in biometrics systems are pointed out.