6 citations · 12 across the 18 of their papers we have counts for
14 papers · 1 filter
Verified SHAP: Provable Bounds for Exact Shapley Values of Neural Networks
David Boetius, Shahaf Bassan, Guy Katz +2
Shapley additive explanations (SHAP) are widely recognised as computationally intractable for neural networks, since they induce an exponential search space over the input features…
Unifying Formal Explanations: A Complexity-Theoretic Perspective
Shahaf Bassan, Xuanxiang Huang, Guy Katz
Previous work has explored the computational complexity of deriving two fundamental types of explanations for ML model predictions: (1) *sufficient reasons*, which are subsets of i…
Provably Explaining Neural Additive Models
Shahaf Bassan, Yizhak Yisrael Elboher, Tobias Ladner +4
Despite significant progress in post-hoc explanation methods for neural networks, many remain heuristic and lack provable guarantees. A key approach for obtaining explanations with…
Formal Mechanistic Interpretability: Automated Circuit Discovery with Provable Guarantees
Itamar Hadad, Guy Katz, Shahaf Bassan
*Automated circuit discovery* is a central tool in mechanistic interpretability for identifying the internal components of neural networks responsible for specific behaviors. While…
SHAP Meets Tensor Networks: Provably Tractable Explanations with Parallelism
Reda Marzouk, Shahaf Bassan, Guy Katz
Although Shapley additive explanations (SHAP) can be computed in polynomial time for simple models like decision trees, they unfortunately become NP-hard to compute for more expres…
Additive Models Explained: A Computational Complexity Approach
Shahaf Bassan, Michal Moshkovitz, Guy Katz
Generalized Additive Models (GAMs) are commonly considered *interpretable* within the ML community, as their structure makes the relationship between inputs and outputs relatively…
Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanations
Shahaf Bassan, Yizhak Yisrael Elboher, Tobias Ladner +2
Despite significant advancements in post-hoc explainability techniques for neural networks, many current methods rely on heuristics and do not provide formally provable guarantees…
What makes an Ensemble (Un) Interpretable?
Shahaf Bassan, Guy Amir, Meirav Zehavi +1
Ensemble models are widely recognized in the ML community for their limited interpretability. For instance, while a single decision tree is considered interpretable, ensembles of t…
Self-Explaining Neural Networks for Business Process Monitoring
Shahaf Bassan, Shlomit Gur, Sergey Zeltyn +3
Tasks in Predictive Business Process Monitoring (PBPM), such as Next Activity Prediction, focus on generating useful business predictions from historical case logs. Recently, Deep…
Explain Yourself, Briefly! Self-Explaining Neural Networks with Concise Sufficient Reasons
Shahaf Bassan, Ron Eliav, Shlomit Gur
*Minimal sufficient reasons* represent a prevalent form of explanation - the smallest subset of input features which, when held constant at their corresponding values, ensure that…
On the Computational Tractability of the (Many) Shapley Values
Reda Marzouk, Shahaf Bassan, Guy Katz +1
Recent studies have examined the computational complexity of computing Shapley additive explanations (also known as SHAP) across various models and distributions, revealing their t…
Hard to Explain: On the Computational Hardness of In-Distribution Model Interpretation
Guy Amir, Shahaf Bassan, Guy Katz
The ability to interpret Machine Learning (ML) models is becoming increasingly essential. However, despite significant progress in the field, there remains a lack of rigorous chara…
Local vs. Global Interpretability: A Computational Complexity Perspective
Shahaf Bassan, Guy Amir, Guy Katz
The local and global interpretability of various ML models has been studied extensively in recent years. However, despite significant progress in the field, many known results rema…
Towards Formal XAI: Formally Approximate Minimal Explanations of Neural Networks
Shahaf Bassan, Guy Katz
With the rapid growth of machine learning, deep neural networks (DNNs) are now being used in numerous domains. Unfortunately, DNNs are "black-boxes", and cannot be interpreted by h…