Interpretation maps with guaranteed robustness

US11341598B2 · kind B2 · utility

Assignees

• International Business Machines Corporation · US
• Rensselaer Polytechnic Institute · US

Inventors

• Ao Liu · Beijing, CN
• Sijia Liu · Beijing, CN
• Abhishek Bhandwaldar · Somerville, US
• Chuang Gan · Cambridge, US
• Lirong Xia · Troy, US
• Qi Cheng Li · Beijing, CN

Key dates

• Filing date: Jun 5, 2020
• Grant date: May 24, 2022
• Expiry date: Jan 21, 2041

Classification

• Technology area (CPC G): Physics
• CPC primary: G06T2207/20216
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Interpretation maps of deep neural networks are provided that use Renyi differential privacy to guarantee the robustness of the interpretation. In one aspect, a method for generating interpretation maps with guaranteed robustness includes: perturbing an original digital image by adding Gaussian noise to the original digital image to obtain m noisy images; providing the m noisy images as input to a deep neural network; interpreting output from the deep neural network to obtain m noisy interpretations corresponding to the m noisy images; thresholding the m noisy interpretations to obtain a top-k of the m noisy interpretations; and averaging the top-k of the m noisy interpretations to produce an interpretation map with certifiable robustness.