Lg Electronics Inc.

University of Toronto

Jongseong Jang

Hyunwoo Kim

Sangmin Lee

Konstantinos Plataniotis

Shervin Mehryar

Sam Sattarzadeh

Mahesh Sudhakar

YeonJeong Jeong

Anthony Lem

Embodiments of the present disclosure relate to generating explanation maps for explaining convolutional neural networks through attribution-based input sampling and block-wise feature aggregation. An example of a disclosed method for generating an explanation map for a convolutional neural network (CNN) includes obtaining an input image resulting in an output determination of the CNN, selecting a plurality of feature maps extracted from a plurality of pooling layers of the CNN, generating a plurality of attribution masks based on the plurality of feature maps, applying the generated attribution masks to the input image to obtain a plurality of visualization maps, and generating an explanation map of the output determination of the CNN based on the plurality of visualization maps.

Semantic input sampling for explanation (SISE) of convolutional neural networks

Growing community of inventors

Toronto, Canada

Jongseong Jang