Robustness (computer science): Difference between revisions

Revision as of 14:27, 9 December 2015

Revision as of 09:31, 11 December 2015

Line 1:

{{mergetomerge to|Fault-tolerant computer system|date=December 2014}}

In [[computer science]], '''robustness''' is the ability of a computer system to cope with errors during execution. Robustness can also be defined as the ability of an algorithm to continue operating despite abnormalities in input, calculations, etc. Robustness can encompass many areas of computer science, such as [[Defensive programming|robust programming]], [[Overfitting|robust machine learning]], and [[Robust Security Network]]. Formal techniques, such as [[fuzz testing]], are essential to showing robustness since this type of testing involves invalid or unexpected inputs. Alternatively, [[fault injection]] can be used to test robustness. Various commercial products perform robustness testing of software systems, and is a process of [[failure assessment]] analysis.<ref>http://www.stanford.edu/~bakerjw/Publications/Baker%20et%20al%20(2008)%20Robustness,%20Structural%20Safety.pdf</ref>

Line 28:

=== Robust machine learning ===

Robust machine learning typically refers to the robustness of machine learning algorithms. For a machine learning algorithm to be considered robust, either the testing error has to be consistent with the training error, or the performance is stable after adding some noise to the dataset.<ref>http://www.researchgate.net/post/What_is_the_definition_of_the_robustness_of_a_machine_learning_algorithm is based on rich language it means it is easy to understand by the user'''''Bold text'''''.</ref>

=== Robust network design ===