Quanshi Zhang

20

Papers

36

Total Citations

Papers (20)

Explaining Generalization Power of a DNN Using Interactive Concepts

Batch Normalization Is Blind to the First and Second Derivatives of the Loss

Layerwise Change of Knowledge in Neural Networks

Mining Object Parts From CNNs via Active Question-Answering

Interpretable Convolutional Neural Networks

Interpreting CNNs via Decision Trees

Explaining Knowledge Distillation by Quantifying the Knowledge

Verifiability and Predictability: Interpreting Utilities of Network Architectures for Point Cloud Processing

Defining and Quantifying the Emergence of Sparse Concepts in DNNs

Mining And-Or Graphs for Graph Matching and Object Discovery

Explaining Neural Networks Semantically and Quantitatively

Interpreting Attributions and Interactions of Adversarial Attacks

3D-Rotation-Equivariant Quaternion Neural Networks

Monitoring Primitive Interactions During the Training of DNNs

Clarifying the Behavior and the Difficulty of Adversarial Training

Towards a Unified Game-Theoretic View of Adversarial Perturbations and Robustness

Visualizing the Emergence of Intermediate Visual Patterns in DNNs

Interpreting Representation Quality of DNNs for 3D Point Cloud Processing

Towards the Difficulty for a Deep Neural Network to Learn Concepts of Different Complexities

Towards a Deep and Unified Understanding of Deep Neural Models in NLP