Dmitry Kazhdan

Towards Robust Metrics for Concept Representation Evaluation

Jan 25, 2023
Mateo Espinosa Zarlenga, Pietro Barbiero, Zohreh Shams, Dmitry Kazhdan, Umang Bhatt, Adrian Weller, Mateja Jamnik

Figure 2 for Towards Robust Metrics for Concept Representation Evaluation

Figure 3 for Towards Robust Metrics for Concept Representation Evaluation

Figure 4 for Towards Robust Metrics for Concept Representation Evaluation

* To appear at AAAI 2023

Access Paper or Ask Questions

Explainer Divergence Scores (EDS): Some Post-Hoc Explanations May be Effective for Detecting Unknown Spurious Correlations

Nov 14, 2022
Shea Cardozo, Gabriel Islas Montero, Dmitry Kazhdan, Botty Dimanov, Maleakhi Wijaya, Mateja Jamnik, Pietro Lio

Figure 2 for Explainer Divergence Scores (EDS): Some Post-Hoc Explanations May be Effective for Detecting Unknown Spurious Correlations

Figure 3 for Explainer Divergence Scores (EDS): Some Post-Hoc Explanations May be Effective for Detecting Unknown Spurious Correlations

Figure 4 for Explainer Divergence Scores (EDS): Some Post-Hoc Explanations May be Effective for Detecting Unknown Spurious Correlations

* Presented at the AIMLAI workshop at the 31st ACM International Conference on Information and Knowledge Management (CIKM 2022)

Access Paper or Ask Questions

Encoding Concepts in Graph Neural Networks

Aug 07, 2022
Lucie Charlotte Magister, Pietro Barbiero, Dmitry Kazhdan, Federico Siciliano, Gabriele Ciravegna, Fabrizio Silvestri, Mateja Jamnik, Pietro Lio

Figure 2 for Encoding Concepts in Graph Neural Networks

Figure 3 for Encoding Concepts in Graph Neural Networks

Figure 4 for Encoding Concepts in Graph Neural Networks

Access Paper or Ask Questions

GCExplainer: Human-in-the-Loop Concept-based Explanations for Graph Neural Networks

Jul 25, 2021
Lucie Charlotte Magister, Dmitry Kazhdan, Vikash Singh, Pietro Liò

Figure 2 for GCExplainer: Human-in-the-Loop Concept-based Explanations for Graph Neural Networks

Figure 3 for GCExplainer: Human-in-the-Loop Concept-based Explanations for Graph Neural Networks

Figure 4 for GCExplainer: Human-in-the-Loop Concept-based Explanations for Graph Neural Networks

* Accepted as 3rd ICML Workshop on Human in the Loop Learning, 2021

Access Paper or Ask Questions

Algorithmic Concept-based Explainable Reasoning

Jul 15, 2021
Dobrik Georgiev, Pietro Barbiero, Dmitry Kazhdan, Petar Veličković, Pietro Liò

Figure 2 for Algorithmic Concept-based Explainable Reasoning

Figure 3 for Algorithmic Concept-based Explainable Reasoning

Figure 4 for Algorithmic Concept-based Explainable Reasoning

* preprint

Access Paper or Ask Questions

Failing Conceptually: Concept-Based Explanations of Dataset Shift

May 01, 2021
Maleakhi A. Wijaya, Dmitry Kazhdan, Botty Dimanov, Mateja Jamnik

Figure 2 for Failing Conceptually: Concept-Based Explanations of Dataset Shift

Figure 3 for Failing Conceptually: Concept-Based Explanations of Dataset Shift

Figure 4 for Failing Conceptually: Concept-Based Explanations of Dataset Shift

* ICLR 2021 Workshop (RobustML), 16 pages, 14 figures; typos corrected

Access Paper or Ask Questions

Manipulating SGD with Data Ordering Attacks

Apr 19, 2021
Ilia Shumailov, Zakhar Shumaylov, Dmitry Kazhdan, Yiren Zhao, Nicolas Papernot, Murat A. Erdogdu, Ross Anderson

Figure 2 for Manipulating SGD with Data Ordering Attacks

Figure 3 for Manipulating SGD with Data Ordering Attacks

Figure 4 for Manipulating SGD with Data Ordering Attacks

Access Paper or Ask Questions

Is Disentanglement all you need? Comparing Concept-based & Disentanglement Approaches

Apr 14, 2021
Dmitry Kazhdan, Botty Dimanov, Helena Andres Terre, Mateja Jamnik, Pietro Liò, Adrian Weller

Figure 2 for Is Disentanglement all you need? Comparing Concept-based & Disentanglement Approaches

Figure 3 for Is Disentanglement all you need? Comparing Concept-based & Disentanglement Approaches

Figure 4 for Is Disentanglement all you need? Comparing Concept-based & Disentanglement Approaches

* Presented at the RAI, WeaSul, and RobustML workshops at The Ninth International Conference on Learning Representations (ICLR) 2021

Access Paper or Ask Questions

MEME: Generating RNN Model Explanations via Model Extraction

Dec 13, 2020
Dmitry Kazhdan, Botty Dimanov, Mateja Jamnik, Pietro Liò

Figure 2 for MEME: Generating RNN Model Explanations via Model Extraction

Figure 3 for MEME: Generating RNN Model Explanations via Model Extraction

Figure 4 for MEME: Generating RNN Model Explanations via Model Extraction

* Presented at the HAMLETS workshop at the 34th Conference on Neural Information Processing Systems (NeurIPS 2020)

Access Paper or Ask Questions

1

2

>>