Timezone: »
Machine learning transparency calls for interpretable explanations of how inputs relate to predictions. Feature attribution is a way to analyze the impact of features on predictions. Feature interactions are the contextual dependence between features that jointly impact predictions. There are a number of methods that extract feature interactions in prediction models; however, the methods that assign attributions to interactions are either uninterpretable, model-specific, or non-axiomatic. We propose an interaction attribution and detection framework called Archipelago which addresses these problems and is also scalable in real-world settings. Our experiments on standard annotation labels indicate our approach provides significantly more interpretable explanations than comparable methods, which is important for analyzing the impact of interactions on predictions. We also provide accompanying visualizations of our approach that give new insights into deep neural networks.
Author Information
Michael Tsang (University of Southern California)
Sirisha Rambhatla (University of Southern California)
Yan Liu (University of Southern California)
More from the Same Authors
-
2022 : A Synthetic Limit Order Book Dataset for Benchmarking Forecasting Algorithms under Distributional Shift »
Defu Cao · Yousef El-Laham · Loc Trinh · Svitlana Vyetrenko · Yan Liu -
2023 Poster: Hierarchical Gaussian Mixture based Task Generative Model for Robust Meta-Learning »
Yizhou Zhang · Jingchao Ni · Wei Cheng · Zhengzhang Chen · Liang Tong · Haifeng Chen · Yan Liu -
2022 Poster: Sparse Interaction Additive Networks via Feature Interaction Detection and Sparse Selection »
James Enouen · Yan Liu -
2022 Poster: Counterfactual Neural Temporal Point Process for Estimating Causal Influence of Misinformation on Social Media »
Yizhou Zhang · Defu Cao · Yan Liu -
2021 Affinity Workshop: WiML Workshop 4 »
Soomin Aga Lee · Meera Desai · Nezihe Merve Gürel · Boyi Li · Linh Tran · Akiko Eriguchi · Jieyu Zhao · Salomey Osei · Sirisha Rambhatla · Geeticka Chauhan · Nwamaka (Amaka) Okafor · Mariya Vasileva -
2021 Affinity Workshop: WiML Workshop 3 »
Soomin Aga Lee · Meera Desai · Nezihe Merve Gürel · Boyi Li · Linh Tran · Akiko Eriguchi · Jieyu Zhao · Salomey Osei · Sirisha Rambhatla · Geeticka Chauhan · Nwamaka (Amaka) Okafor · Mariya Vasileva -
2021 Affinity Workshop: WiML Workshop 2 »
Soomin Aga Lee · Meera Desai · Nezihe Merve Gürel · Boyi Li · Linh Tran · Akiko Eriguchi · Jieyu Zhao · Salomey Osei · Sirisha Rambhatla · Geeticka Chauhan · Nwamaka (Amaka) Okafor · Mariya Vasileva -
2021 Poster: VigDet: Knowledge Informed Neural Temporal Point Process for Coordination Detection on Social Media »
Yizhou Zhang · Karishma Sharma · Yan Liu -
2021 Affinity Workshop: WiML Workshop 1 »
Soomin Aga Lee · Meera Desai · Nezihe Merve Gürel · Boyi Li · Linh Tran · Akiko Eriguchi · Jieyu Zhao · Salomey Osei · Sirisha Rambhatla · Geeticka Chauhan · Nwamaka (Amaka) Okafor · Mariya Vasileva -
2020 Poster: Multi-agent Trajectory Prediction with Fuzzy Query Attention »
Nitin Kamra · Hao Zhu · Dweep Trivedi · Ming Zhang · Yan Liu -
2020 Poster: Provable Online CP/PARAFAC Decomposition of a Structured Tensor via Dictionary Learning »
Sirisha Rambhatla · Xingguo Li · Jarvis Haupt -
2019 Workshop: Learning with Temporal Point Processes »
Manuel Rodriguez · Le Song · Isabel Valera · Yan Liu · Abir De · Hongyuan Zha -
2018 Poster: Neural Interaction Transparency (NIT): Disentangling Learned Interactions for Improved Interpretability »
Michael Tsang · Hanpeng Liu · Sanjay Purushotham · Pavankumar Murali · Yan Liu -
2017 : Posters 1 »
J.P. Lewis · Housam Khalifa Bashier Babiker · Zhongang Qi · Laura Rieger · Ning Xie · Filip Dabek · Koushik Nagasubramanian · Bolei Zhou · Dieuwke Hupkes · CHUN-HAO CHANG · Pamela K Douglas · Enea Ceolini · Derek Doran · Yan Liu · Fuxin Li · Randolph Goebel -
2017 : Methods 1 »
Grégoire Montavon · Michael Tsang · Marco Ancona -
2016 Workshop: Learning with Tensors: Why Now and How? »
Anima Anandkumar · Rong Ge · Yan Liu · Maximilian Nickel · Qi (Rose) Yu -
2016 Poster: SPALS: Fast Alternating Least Squares via Implicit Leverage Scores Sampling »
Dehua Cheng · Richard Peng · Yan Liu · Kimis Perros -
2016 Poster: Learning Influence Functions from Incomplete Observations »
Xinran He · Ke Xu · David Kempe · Yan Liu -
2014 Poster: Fast Multivariate Spatio-temporal Analysis via Low Rank Tensor Learning »
Mohammad Taha Bahadori · Qi (Rose) Yu · Yan Liu -
2014 Spotlight: Fast Multivariate Spatio-temporal Analysis via Low Rank Tensor Learning »
Mohammad Taha Bahadori · Qi (Rose) Yu · Yan Liu