Paper Title
Data Poisoning in Supervised Machine Learning: A Comprehensive Survey of Attacks and Defences

Abstract
Machine Learning (ML) is now ubiquitous in all fields but is becoming more vulnerable to data poisoning attacks, wherein the training data is tampered with by the adversaries to defeat the learning process. This survey surveys the state of such attacks, defence mechanisms, and calls for standardized benchmarking and realistic attack simulations for building better security of supervised learning systems. Keywords - Data Poisoning Attacks, Machine Learning, Adversarial Attacks, Defence Mechanisms, Adversarial Machine Learning.