Interpreting Multivariate Shapley Interactions in DNNs

02/02/2021

Interpreting Multivariate Shapley Interactions in DNNs

Hao Zhang, Yichen Xie, Longjie Zheng, Die Zhang, Quanshi Zhang

Keywords:

Abstract Paper Similar Papers

Abstract: This paper aims to explain deep neural networks (DNNs) from the perspective of multivariate interactions. In this paper, we define and quantify the significance of interactions among multiple input variables of the DNN. Input variables with strong interactions usually form a coalition and reflect prototype features, which are memorized and used by the DNN for inference. We define the significance of interactions based on the Shapley value, which is designed to assign the attribution value of each input variable to the inference. We have conducted experiments with various DNNs. Experimental results have demonstrated the effectiveness of the proposed method.

The video of this talk cannot be embedded. You can watch it here:

https://slideslive.com/38949214

(Link will open in new window)

0

0

0

0

Share

This is an embedded video. Talk and the respective paper are published at AAAI 2021 virtual conference. If you are one of the authors of the paper and want to manage your upload, see the question "My papertalk has been externally embedded..." in the FAQ section.

Comments

Post Comment

no comments yet

Similar Papers

06/12/2020

On 1/n neural representation and robustness

Josue Nassar, Piotr Sokol, Sueyeon Chung and
Daniel D Harris, Memming Park

Keywords Paper

0

0

0

0

3:14

07/09/2020

Feature Binding with Category-Dependant MixUp for Semantic Segmentation and Adversarial Robustness

Md Amirul Islam, Matthew Kowal, Konstantinos Derpanis, Neil D. B. Bruce

Keywords Paper

Image Blending, Mixup, Feature Binding, Semantic Segmentation, Adversarial Robustness

0

0

0

0

9:00

18/07/2021

LieTransformer: Equivariant Self-Attention for Lie Groups

Michael Hutchinson, Charline Le Lan, Sheheryar Zaidi and
Emilien Dupont, Yee-Whye Teh, Hyunjik Kim

Keywords Paper

Deep Learning

0

0

0

0

5:18

06/12/2021

Deep Neural Networks as Point Estimates for Deep Gaussian Processes

Vincent Dutordoir, James Hensman, Mark van der Wilk and
Carl Henrik Ek, Zoubin Ghahramani, Nicolas Durrande

Keywords Paper

theory, deep learning, machine learning, generative model, kernel methods

0

0

0

0

14:00

19/08/2021

Information-Theoretic Methods in Deep Neural Networks: Recent Advances and Emerging Opportunities

Shujian Yu, Luis Sanchez Giraldo, Jose Principe

Keywords Paper

Machine learning, General

0

0

0

0

14:07

06/12/2020

How does This Interaction Affect Me? Interpretable Attribution for Feature Interactions

Michael Tsang, Sirisha Rambhatla, Yan Liu

Keywords Paper

0

0

0

0

3:12

06/12/2020

Detecting Interactions from Neural Networks via Topological Analysis

Liu Liu, Qingquan Song, Kaixiong Zhou and
Ting-Hsiang Wang, Ying Shan, Xia Hu

Keywords Paper

Algorithms -> Bandit Algorithms; Reinforcement Learning and Planning -> Reinforcement Learning; Theory -> Learning Theory, Reinforcement Learning and Planning -> Exploration

0

0

0

0

3:25

26/04/2020

A Constructive Prediction of the Generalization Error Across Scales

Jonathan S. Rosenfeld, Amir Rosenfeld, Yonatan Belinkov, Nir Shavit

Keywords Paper

neural networks, deep learning, generalization error, scaling, scalability, vision, language

0

0

0

0

4:59

13/04/2021

Influence decompositions for neural network attribution

Kyle Reing, Greg Ver Steeg, Aram Galstyan

Keywords Paper

0

0

0

0

2:52

18/07/2021

On the Implicit Bias of Initialization Shape: Beyond Infinitesimal Mirror Descent

Shahar Azulay, Edward Moroshko, Mor Shpigel Nacson and
Blake Woodworth, Nati Srebro, Amir Globerson, Daniel Soudry

Keywords Paper

, Probabilistic Methods, MCMC, Theory, Deep learning Theory

0

0

0

0

15:38

12/07/2020

The Buckley-Osthus model and the block preferential attachment model: statistical analysis and application

Wenpin Tang, Xin Guo, Fengmin Tang

Keywords Paper

Sequential, Network, and Time-Series Modeling

0

0

0

0

15:18

23/08/2020

Controllable multi-interest framework for recommendation

Yukuo Cen, Jianwei Zhang, Xu Zou and
Chang Zhou, Hongxia Yang, Jie Tang

Keywords Paper

recommender system, multi-interest framework, sequential recommendation

0

0

0

0

15:59

14/09/2020

A³ : Activation Anomaly Analysis

Philip Sperl, Jan-Philipp Schulze, Konstantin Böttinger

Keywords Paper

anomaly detection, deep learning, intrusion detection, semi-supervised learning, coverage analysis, data mining, it security

0

0

0

0

15:21

12/07/2020

Information-Theoretic Local Minima Characterization and Regularization

Zhiwei Jia, Hao Su

Keywords Paper

Deep Learning - Theory

0

0

0

0

14:11

06/12/2021

Row-clustering of a Point Process-valued Matrix

Lihao Yin, Ganggang Xu, Huiyan Sang, Yongtao Guan

Keywords Paper

machine learning, clustering

0

0

0

0

14:30

26/04/2020

Sharing Knowledge in Multi-Task Deep Reinforcement Learning

Carlo D'Eramo, Davide Tateo, Andrea Bonarini and
Marcello Restelli, Jan Peters

Keywords Paper

Deep Reinforcement Learning, Multi-Task

0

0

0

0

4:27

06/12/2021

On Inductive Biases for Heterogeneous Treatment Effect Estimation

Alicia Curth, Mihaela van der Schaar

Keywords Paper

deep learning, causality

0

0

0

0

13:17

19/08/2021

Compositional Neural Logic Programming

Son N. Tran

Keywords Paper

Machine Learning, Neuro-Symbolic Methods, Leveraging Knowledge and Learning

0

0

0

0

10:48

26/04/2020

Theory and Evaluation Metrics for Learning Disentangled Representations

Kien Do, Truyen Tran

Keywords Paper

disentanglement, metrics

0

0

0

0

3:37

06/12/2020

A Causal View on Robustness of Neural Networks

Cheng Zhang, Kun Zhang, Yingzhen Li

Keywords Paper

Data, Challenges, Implementations, and Software -> Virtual Environments; Deep Learning -> Memory-Augmented Neural Networks; Neu, Deep Learning

0

0

0

0

3:25

06/12/2020

Multi-task Causal Learning with Gaussian Processes

Virginia Aglietti, Theo Damoulas, Mauricio Álvarez, Javier González

Keywords Paper

0

0

0

0

3:14

30/11/2020

Sequential View Synthesis with Transformer

Phong Nguyen-Ha, Lam Huynh, Esa Rahtu, Janne Heikkila

Keywords Paper

0

0

0

0

9:38

13/04/2021

Deep fourier kernel for self-attentive point processes

Shixiang Zhu, Minghe Zhang, Ruyi Ding, Yao Xie

Keywords Paper

0

0

0

0

3:08

18/07/2021

What's in the Box? Exploring the Inner Life of Neural Networks with Robust Rules

Jonas Fischer, Anna Olah, Jilles Vreeken

Keywords Paper

Deep Learning

0

0

0

0

5:34

06/12/2020

Adversarial Training is a Form of Data-dependent Operator Norm Regularization

Kevin Roth, Yannic Kilcher, Thomas Hofmann

Keywords Paper

0

0

0

0

3:20

06/12/2020

Network Diffusions via Neural Mean-Field Dynamics

shushan He, Hongyuan Zha, Xiaojing Ye

Keywords Paper

0

0

0

0

3:21

02/06/2020

Entity Summarization with User Feedback

Qingxia Liu, Yue Chen, Gong Cheng and
Evgeny Kharlamov, Junyou Li, Yuzhong Qu

Keywords Paper

0

0

0

0

21:30

13/04/2021

DebiNet: Debiasing linear models with nonlinear overparameterized neural networks

Shiyun Xu, Zhiqi Bu

Keywords Paper

0

0

0

0

2:56

12/07/2020

Doubly Stochastic Variational Inference for Neural Processes with Hierarchical Latent Variables

Qi Wang, Herke van Hoof

Keywords Paper

Deep Learning - Generative Models and Autoencoders

0

0

0

0

14:18

16/11/2020

Neural Conversational QA: Learning to Reason vs Exploiting Patterns

Nikhil Verma, Abhishek Sharma, Dhiraj Madan and
Danish Contractor, Harshit Kumar, Sachindra Joshi

Keywords Paper

neural tasks, sharc task, sharc, heuristic-based program

0

0

0

0

7:04

03/05/2021

The Recurrent Neural Tangent Kernel

Sina Alemohammad, Jack Wang, Randall Balestriero, Richard Baraniuk

Keywords Paper

Gaussian Process, Recurrent Neural Network, Neural Tangent Kernel, Overparameterization

0

0

0

0

4:44

06/12/2021

Framing RNN as a kernel method: A neural ODE approach

Adeline Fermanian, Pierre Marion, Jean-Philippe Vert, Gérard Biau

Keywords Paper

theory, deep learning, kernel methods

0

0

0

0

18:52

02/02/2021

Liquid Time-constant Networks

Ramin Hasani, Mathias Lechner, Alexander Amini and
Daniela Rus, Radu Grosu

Keywords Paper

0

0

0

0

16:02

18/07/2021

Rethinking Neural vs. Matrix-Factorization Collaborative Filtering: the Theoretical Perspectives

Da Xu, Chuanwei Ruan, Evren Korpeoglu and
Sushant Kumar, Kannan Achan

Keywords Paper

Algorithms, Algorithms, Structured Prediction, Algorithms, Collaborative Filtering

0

0

0

0

5:14

02/02/2021

Shuffling Recurrent Neural Networks

Michael Rotman, Lior Wolf

Keywords Paper

0

0

0

0

20:10

12/07/2020

A Flexible Latent Space Model for Multilayer Networks

Xuefei Zhang, Songkai Xue, Ji Zhu

Keywords Paper

Sequential, Network, and Time-Series Modeling

0

0

0

0

15:40

06/12/2021

Periodic Activation Functions Induce Stationarity

Lassi Meronen, Martin Trapp, Arno Solin

Keywords Paper

deep learning, kernel methods

0

0

0

0

12:24

18/07/2021

Robust Learning for Data Poisoning Attacks

Yunjuan Wang, Poorya Mianjy, Raman Arora

Keywords Paper

Deep Learning, Generative Models, Algorithms, Unsupervised Learning; Deep Learning, Adversarial Networks, Algorithms, Adversarial Examples

0

0

0

0

5:20

22/11/2021

Probabilistic Regression with Huber Distributions

David A Mohlin, Gerald Bianchi, Josephine Sullivan

Keywords Paper

regression, probabilities, uncertainty, body pose, facial landmarks, huber loss, robust estimation

0

0

0

0

2:31

03/08/2020

Prediction Intervals: Split Normal Mixture from Quality-Driven Deep Ensembles

Tárik S. Salem, Helge Langseth, Heri Ramampiaro

Keywords Paper

0

0

0

0

7:45