Benchmarking the Effect of Poisoning Defenses on the Security and Bias of Deep Learning Models | IEEE Conference Publication | IEEE Xplore