סילבוס Syllabus Силабус خطة الدورة

The seminar will study the concept of “explanation” in databases. Why did I get a specific answer? Why didn’t I get another answer that I expected to see? Which of the database tuples is responsible to the answers or non-answers? Why is the database inconsistent, that is, violates integrity constraints? Which of the tuples is responsible to the inconsistency? How do you measure the level of responsibility? What is the computational complexity of calculating different responsibility measures, exactly or approximately? How do you visualize complex data in a comprehensible manner? What techniques are used by systems to provide explanations effectively?  In the seminar, we will study articles on the topic, some new and some older; in particular, we will learn how causality theory and game theory are used for defining explanations in a general manner that is domain independent.

Lecture Planning

<google-sheets-html-origin style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);"></google-sheets-html-origin>

<colgroup><col width="288"><col width="168"><col width="100"><col width="100"></colgroup>

Weeks 3-4 (2 students):
Scorpion: Explaining Away Outliers in Aggregate Queries	Eugene Wu, Samuel Madden	Scorpion: Explaining Away Outliers in Aggregate Queries	VLDB 2013
Optimal Histograms with Outliers	Rachel Behar ; Sara Cohen	Optimal Histograms with Outliers	EDBT 2020
Interpretable and Informative Explanations of Outcomes	El Gebaly, Kareem, Parag Agrawal, Lukasz Golab, Flip Korn, Divesh Srivastava	Interpretable and Informative Explanations of Outcomes	VLDB 2014
Week 5-6 (3 students):
The Complexity of Causality and Responsibility for Query Answers and non-Answers.	Alexandra Meliou, Wolfgang Gatterbauer, Katherine F. Moore, Dan Suciu	The Complexity of Causality and Responsibility for Query Answers and non-Answers	VLDB 2010
A Formal Approach to Finding Explanations for Database Queries	Sudeepa Roy ; Dan Suciu	A formal approach to finding explanations for database queries | Proceedings of the 2014 ACM SIGMOD International Conference on Management of Data	SIGMOD 2014
Approximate Summaries for Why and Why-not Provenance.	Seokki Lee, Bertram Ludäscher, Boris Glavic	Approximate Summaries for Why and Why-not Provenance	VLDB 2020
Week 7-8 (3 students):
The Shapley Value of Tuples in Query Answering	Livshits, Ester ; Bertossi, Leopoldo ; Kimelfeld, Benny ; Sebag, Moshe	The Shapley Value of Tuples in Query Answering	ICDT 2020
Computing the Shapley Value of Facts in Query Answering	Daniel Deutch, Nave Frost, Benny Kimelfeld, Mikaël Monet	Computing the Shapley Value of Facts in Query Answering | Proceedings of the 2022 International Conference on Management of Data	SIGMOD 2022
Week 9-10 (3 students):
A Unified Approach to Interpreting Model Predictions	Scott M. Lundberg, Su-In Lee	A Unified Approach to Interpreting Model Predictions	NEURIPS 2017
“Why Should I Trust You?”: Explaining the Predictions of Any Classifier	Marco Tulio Ribeiro, Sameer Singh, Carlos Guestrin	“Why Should I Trust You?” Explaining the Predictions of Any Classifier	KDD 2016
Anchors: High-Precision Model-Agnostic Explanations	Marco Túlio Ribeiro, Sameer Singh, Carlos Guestrin	Anchors: High-Precision Model-Agnostic Explanations | Proceedings of the AAAI Conference on Artificial Intelligence	AAAI 2018
Weeks 11-12 (3 students):
Properties of Inconsistency Measures for Databases	Ester Livshits, Rina Kochirgan, Segev Tsur, Ihab F. Ilyas, Benny Kimelfeld, Sudeepa Roy	Properties of Inconsistency Measures for Databases | Proceedings of the 2021 International Conference on Management of Data
The Shapley value of inconsistency measures for functional dependencies	Ester Livshits, Benny Kimelfeld	The Shapley Value of Inconsistency Measures for Functional Dependencies	CIKM 2021
Explanations for Data Repair Through Shapley Values	Daniel Deutch, Nave Frost, Amir Gilad, Oren Sheffer	Explanations for Data Repair Through Shapley Values | Proceedings of the 30th ACM International Conference on Information & Knowledge Management