Search results
Jan 8, 2018 · The framework categorizes the suitability of metrics to a decision-maker based on (1) the decision-context (i.e., the suitability of using absolute performance or regret), (2) the decision-maker's preferred level of risk aversion, and (3) the decision-maker's preference toward maximizing performance, minimizing variance, or some higher-order mom...
Current methods for training robust networks lead to a drop in test accuracy, which has led prior works to posit that a robustness-accuracy tradeoff may be inevitable in deep learning. We take a closer look at this phenomenon and first show that real image datasets are actually separated. With this property in mind, we then prove
6 days ago · The first big difference is that you calculate accuracy on the predicted classes while you calculate ROC AUC on predicted scores. That means you will have to find the optimal threshold for your problem. Moreover, accuracy looks at fractions of correctly assigned positive and negative classes.
Jan 8, 2018 · To address this shortcoming, an approach is developed that uses metamodels (surrogates of computationally expensive simulation models) to calculate robustness and other objectives.
It is possible to find both robust and accurate model for D-separated data. Consider function ’:#→ℝ"and !∈#with true label &∈*, if . I.’is + / -Locally Lipschitz in radius Daround !, and. II. ,−’! (≥2for all O≠& Then >!=argmin. #is astute at !with radius D. Intuitively, Condition-I indicates that the changes of prediction in 4!,Dare slow.
Jan 8, 2018 · The framework categorizes the suitability of metrics to a decision-maker based on (1) the decision-context (i.e., the suitability of using absolute performance or regret), (2) the decision-maker's preferred level of risk aversion, and (3) the decision-maker's preference toward maximizing performance, minimizing variance, or some higher-order mom...
A Closer Look at Accuracy vs. Robustness. Review 1. Summary and Contributions: The paper examines robustness-accuracy tradeoffs, extending previous work of Chaudhari’s group on robustness (measured by astuteness) and local smoothness examined by multiple groups. The theoretical results are fairly restrictive, but the empirical results add ...
Current methods for training robust networks lead to a drop in test accuracy, which has led prior works to posit that a robustness-accuracy tradeoff may be inevitable in deep learning. We take a closer look at this phenomenon and first show that real image datasets are actually separated. With this property in mind, we then prove
Jul 11, 2019 · Science, Tech, Math › Math. Robustness in Statistics. Jekaterina Nikitina/Getty Images. By. Courtney Taylor. Updated on July 11, 2019. In statistics, the term robust or robustness refers to the strength of a statistical model, tests, and procedures according to the specific conditions of the statistical analysis a study hopes to achieve.
Our method explores how generated data can be used to improve robust accuracy by +8.96% without us-ing any additional external data. This constitutes the largest jump in robust accuracy in this setting. Our best model reaches a robust accuracy of 66.10% against AA+MT [30]. Figure 2: Overview of our approach.
