You are using an outdated browser. Please upgrade your browser to improve your experience and security.
RobustART is the first comprehensive Robust ness investigation benchmark on large-scale dataset ImageNet regarding AR chitectural design ( 49 human-designed off-the-shelf architectures and 1200+ neural architecture searched networks) and T raining techniques ( 10+ general ones e.g. , extra training data, etc) towards diverse noises (adversarial, natural, and system noises). Our benchmark (including open-source toolkit, pre-trained model zoo, datasets, and analyses): (1) presents an open-source platform for
Thousands of architectures (49 human-designed off-the-shelf architectures and 1200 neural architecture searched networks) and training techniques (10+ general ones)