Robustness definition

Robustness means the property of mitigation measures resulting from combining the safety gain provided by the mitigation measures and the level of assurance and integrity that the safety gain has been achieved;

Split View

AI-Powered Contracts

Draft, Review & Redline at the Speed of AI

Get StartedSchedule a Demo

Robustness. If no player is corrupted during the protocol then all players accept.

Robustness in this context means that the estimator is robust against deviations of the data from the assumed statistical distribution, usually the normal distributi- on. For example, this is not true for least-squares estimators. As a result, in contrast to ordinary estimators such as least squares, robust estimators are not distorted by influential outliers.

Examples of Robustness in a sentence

• Robustness and tamper protections Widevine agreements with device manufacturers include the robustness rules below.

• If an Adopter has (a) actual notice of New Circumstances, or (b) actual knowledge of New Circumstances (the occurrence of (a) or (b) hereinafter referred to as “Notice”), then within eighteen (18) months after Notice such Adopter shall cease distribution of such Licensed Product and shall only distribute Licensed Products that are compliant with the Robustness Rules in view of the then-current circumstances.

• For the purposes of these Robustness Rules, "Software" shall mean the implementation of the content protection requirements as to which this Agreement requires a Licensed Product to be compliant through any computer program code consisting of instructions or data, other than such instructions or data that are included in Hardware.

• If CPUG so requests via DTLA, Adopter shall provide, once per model or version of product, any publicly available technical design documentation and, under a reasonable, mutually-acceptable non- disclosure agreement, the service manual for such product, in order to assist in the evaluation of the compliance of such product with these Robustness Rules.

• Although an implementation of a Licensed Product when designed and first shipped may meet the above standards, subsequent circumstances may arise which, had they existed at the time of design of a particular Licensed Product, would have caused such products to fail to comply with these Robustness Rules (“New Circumstances”).

More Definitions of Robustness

Robustness means the ability to respond to abnormal inputs and conditions.

Robustness means the degree to which software can function when it receives unexpected inputs or within stressful conditions

Robustness means the ability of an AWP system to address a broad variety of constituents and changes in the

Robustness means that an AI system is capable of recognizing and behav- ing correctly when exposed to different scenarios compared to those in which it was trained.195 Also, AI systems must be able to “adequately deal with errors or inconsistencies during all life cycle phases,” and they should be resilient against attacks and attempts to manipulate data and algorithms.196 A further common feature in recent policies is the requirement that the system is not biased, i.e., that the system is fair or equitable.197 The non- discriminatory nature of the system’s outcomes may be verified during the certification stage. As discussed above, since the requirement is tightly linked to data, some guidelines advise to ensure during the training phase that the

Robustness means the ability of the analytical procedure to be resistant to the influence of small specified changes in the test conditions, which indicates its reliability under normal (standard) use.

Robustness. The generated PIs should be robust against variations and uncertainties exhibited by the input biomet- ric data. This enables PIC to compare PIs directly. The applied template protection scheme should not degrade the recognition performance of the biometric system. • Renewability: It should be computationally difficult to obtain the original biometric template from multiple instances PIs derived from the same biometric trait of an individual. This makes it possible to revoke and reissue new instances of compromised PI. • Unlinkability: It should be computationally difficult to know whether two or more instances of PIs were gener- ated from the same biometric trait of a user. This prevents cross-matching across different applications. In this work, to deal with template protection schemes in mobile environments, we analyse most of the template protection schemes and settle on the one deemed suitable for such platform. Considering mobile platform constraints, we focus on bloom filter template protection schemes.

Robustness means the ability of an AWP system to address a broad variety of (i) constituents and (ii) changes in the concentrations of the constituents in the source water and resist a failure.