Generalised Lipschitz Regularisation Equals Distributional Robustness
Cranko, Zac, Shi, Zhan, Zhang, Xinhua, Nock, Richard, Kornblith, Simon
The problem of adversarial examples has highlighted the need for a theory of regularisation that is general enough to apply to exotic function classes, such as universal approximators. In response, we give a very general equality result regarding the relationship between distributional robustness and regularisation, as defined with a transportation cost uncertainty set. The theory allows us to (tightly) certify the robustness properties of a Lipschitz-regularised model with very mild assumptions. As a theoretical application we show a new result explicating the connection between adversarial learning and distributional robustness. We then give new results for how to achieve Lipschitz regularisation of kernel classifiers, which are demonstrated experimentally.
Feb-10-2020
- Country:
- Oceania > Australia
- New South Wales > Sydney (0.04)
- North America
- United States
- New York > New York County
- New York City (0.04)
- Massachusetts
- Suffolk County > Boston (0.04)
- Middlesex County > Cambridge (0.04)
- Louisiana > Orleans Parish
- New Orleans (0.04)
- Illinois > Cook County
- Chicago (0.04)
- Hawaii > Honolulu County
- Honolulu (0.04)
- California
- Los Angeles County > Long Beach (0.04)
- Santa Clara County > San Jose (0.04)
- New York > New York County
- Canada > Quebec
- Montreal (0.04)
- United States
- Europe
- Germany > Berlin (0.04)
- Switzerland > Basel-City
- Basel (0.04)
- Asia > Middle East
- Oceania > Australia
- Genre:
- Research Report > New Finding (0.54)
- Technology: