Publications

(2021). Discovering and Explaining the Representation Bottleneck of DNNs. arxiv 2021.

PDF Source Document

(2021). A General Taylor Framework for Unifying and Revisiting Attribution Methods. arxiv 2021.

PDF

(2021). Mutual Information Preserving Back-propagation Learn to Invert for Faithful Attribution. KDD 2021.

PDF

(2021). A Unified Taylor Framework for Revisiting Attribution Methods. AAAI 2021.

PDF

(2020). Invariant subspace learning for time series data based on dynamic time warping distance. PR 2020.

PDF