Representation Based Complexity Measures for Predicting Generalization in Deep Learning

1 code implementation • 4 Dec 2020 • Parth Natekar, Manik Sharma

Deep Neural Networks can generalize despite being significantly overparametrized.

Abstracting Deep Neural Networks into Concept Graphs for Concept Level Interpretability

1 code implementation • 14 Aug 2020 • Avinash Kori, Parth Natekar, Ganapathy Krishnamurthi, Balaji Srinivasan

Extracting such a graphical representation of the model's behavior on an abstract, higher conceptual level would unravel the learnings of these models and would help us to evaluate the steps taken by the model for predictions.

Brain Tumor Segmentation Decision Making +2

Demystifying Brain Tumour Segmentation Networks: Interpretability and Uncertainty Analysis

1 code implementation • 3 Sep 2019 • Parth Natekar, Avinash Kori, Ganapathy Krishnamurthi

In this paper, we explore various techniques to explain the functional organization of brain tumor segmentation models and to extract visualizations of internal concepts to understand how these networks achieve highly accurate tumor segmentations.

Brain Tumor Segmentation Medical Diagnosis +2