About

Adnan Darwiche is a professor of computer science at UCLA Samueli School of Engineering. His research interests include probabilistic and logical reasoning and their applications in diagnosis, planning, and system design and analysis. He holds an M.S. (1989) and a Ph.D. (1993) in computer science from Stanford University. Darwiche has been recognized as an AAAI Fellow and serves as the Editor-in-Chief of the Journal of Artificial Intelligence Research (JAIR). His work focuses on advancing the understanding and development of reasoning systems that integrate probabilistic and logical methods, contributing significantly to the fields of artificial intelligence and computational reasoning.

Research topics

• Artificial Intelligence
• Computer Science
• Algorithm
• Theoretical computer science
• Mathematics
• Arithmetic
• Machine Learning
• Data Mining
• Discrete mathematics
• Psychology
• Social psychology

Selected publications

• On Explaining Classifiers using Instance Abstractions

  Topoi · 2026-04-26

  articleOpen accessSenior author

  Two types of explanations for the decisions made by classifiers have been studied extensively in the AI literature. The first type explains why a decision was made and is known as a sufficient reason for the decision. The second type explains how a decision can be changed and is known as a necessary reason for the decision. Earlier results showed that these two types of explanations correspond to two particular syntactic forms of the complete reason, which is a condition on the classifier’s input (i.e., instance) that is both sufficient and necessary for the decision. More recently, we showed that when non-binary features are present, a relaxation of the complete reason, called the general reason, contains more information about the decision and the underlying classifier and can be used to formulate explanations that generalize sufficient and necessary reasons. We provide a reconstruction and summary of these earlier results that is founded on viewing the complete and general reasons as “instance abstractions.” We also study the general-reason abstraction further, showing how it can provide information about decisions and classifiers beyond what was identified earlier. Our treatment will give rise to two pillars for modern work on explainability in AI: sufficient and necessary logical conditions which played a central role in certain formulations in philosophy, and minimized syntactic forms (of a condition) which have been developed mainly in computer science.

  PublisherOA PDFDOI

• On the Granularity of Causal Effect Identifiability

  ArXiv.org · 2025-10-19

  preprintOpen accessSenior author

  The classical notion of causal effect identifiability is defined in terms of treatment and outcome variables. In this paper, we consider the identifiability of state-based causal effects: how an intervention on a particular state of treatment variables affects a particular state of outcome variables. We demonstrate that state-based causal effects may be identifiable even when variable-based causal effects may not. Moreover, we show that this separation occurs only when additional knowledge -- such as context-specific independencies -- is available. We further examine knowledge that constrains the states of variables, and show that such knowledge can improve both variable-based and state-based identifiability when combined with other knowledge such as context-specific independencies. We finally propose an approach for identifying causal effects under these additional constraints, and conduct empirical studies to further illustrate the separations between the two levels of identifiability.

  PublisherOA PDFDOI

• Causal Unit Selection using Tractable Arithmetic Circuits

  arXiv (Cornell University) · 2024-04-10

  preprintOpen accessSenior author

  The unit selection problem aims to find objects, called units, that optimize a causal objective function which describes the objects' behavior in a causal context (e.g., selecting customers who are about to churn but would most likely change their mind if encouraged). While early studies focused mainly on bounding a specific class of counterfactual objective functions using data, more recent work allows one to find optimal units exactly by reducing the causal objective to a classical objective on a meta-model, and then applying a variant of the classical Variable Elimination (VE) algorithm to the meta-model -- assuming a fully specified causal model is available. In practice, however, finding optimal units using this approach can be very expensive because the used VE algorithm must be exponential in the constrained treewidth of the meta-model, which is larger and denser than the original model. We address this computational challenge by introducing a new approach for unit selection that is not necessarily limited by the constrained treewidth. This is done through compiling the meta-model into a special class of tractable arithmetic circuits that allows the computation of optimal units in time linear in the circuit size. We finally present empirical results on random causal models that show order-of-magnitude speedups based on the proposed method for solving unit selection.

  PublisherOA PDFDOI

• Identifying Causal Effects Under Functional Dependencies

  arXiv (Cornell University) · 2024-03-07

  preprintOpen accessSenior author

  We study the identification of causal effects, motivated by two improvements to identifiability which can be attained if one knows that some variables in a causal graph are functionally determined by their parents (without needing to know the specific functions). First, an unidentifiable causal effect may become identifiable when certain variables are functional. Second, certain functional variables can be excluded from being observed without affecting the identifiability of a causal effect, which may significantly reduce the number of needed variables in observational data. Our results are largely based on an elimination procedure which removes functional variables from a causal graph while preserving key properties in the resulting causal graph, including the identifiability of causal effects.

  PublisherOA PDFDOI

• Identifying Causal Effects Under Functional Dependencies

  Entropy · 2024-12-06 · 2 citations

  articleOpen accessSenior author

  We study the identification of causal effects, motivated by two improvements to identifiability that can be attained if one knows that some variables in a causal graph are functionally determined by their parents (without needing to know the specific functions). First, an unidentifiable causal effect may become identifiable when certain variables are functional. Secondly, certain functional variables can be excluded from being observed without affecting the identifiability of a causal effect, which may significantly reduce the number of needed variables in observational data. Our results are largely based on an elimination procedure that removes functional variables from a causal graph while preserving key properties in the resulting causal graph, including the identifiability of causal effects. Our treatment of functional dependencies in this context mandates a formal, systematic, and general treatment of positivity assumptions, which are prevalent in the literature on causal effect identifiability and which interact with functional dependencies, leading to another contribution of the presented work.

  PublisherDOI

• Causal Unit Selection using Tractable Arithmetic Circuits

  Proceedings of the ... International Florida Artificial Intelligence Research Society Conference · 2024-05-12

  articleOpen accessSenior author

  The unit selection problem aims to find objects, called units, that optimize a causal objective function which describes the objects' behavior in a causal context (e.g., selecting customers who are about to churn but would most likely change their mind if encouraged). While early studies focused mainly on bounding a specific class of counterfactual objective functions using data, more recent work allows one to find optimal units exactly by reducing the causal objective to a classical objective on a meta-model, and then applying a variant of the classical Variable Elimination (VE) algorithm to the meta-model---assuming a fully specified causal model is available. In practice, however, finding optimal units using this approach can be very expensive because the used VE algorithm must be exponential in the constrained treewidth of the meta-model, which is larger and denser than the original model. We address this computational challenge by introducing a new approach for unit selection that is not necessarily limited by the constrained treewidth. This is done through compiling the meta-model into a special class of tractable arithmetic circuits that allows the computation of optimal units in time linear in the circuit size. We finally present empirical results on random causal models that show order-of-magnitude speedups based on the proposed method for solving unit selection.

  PublisherOA PDFDOI

• Constrained Identifiability of Causal Effects

  arXiv (Cornell University) · 2024-12-03

  preprintOpen accessSenior author

  We study the identification of causal effects in the presence of different types of constraints (e.g., logical constraints) in addition to the causal graph. These constraints impose restrictions on the models (parameterizations) induced by the causal graph, reducing the set of models considered by the identifiability problem. We formalize the notion of constrained identifiability, which takes a set of constraints as another input to the classical definition of identifiability. We then introduce a framework for testing constrained identifiability by employing tractable Arithmetic Circuits (ACs), which enables us to accommodate constraints systematically. We show that this AC-based approach is at least as complete as existing algorithms (e.g., do-calculus) for testing classical identifiability, which only assumes the constraint of strict positivity. We use examples to demonstrate the effectiveness of this AC-based approach by showing that unidentifiable causal effects may become identifiable under different types of constraints.

  PublisherOA PDFDOI

• Towards an effective practice of learning from data and knowledge

  International Journal of Approximate Reasoning · 2024-04-05 · 3 citations

  articleOpen accessSenior authorCorresponding

  We discuss some recent advances on combining data and knowledge in the context of supervised learning using Bayesian networks. A first set of advances concern the computational efficiency of learning and inference, and they include a software-level boost based on compiling Bayesian network structures into tractable circuits in the form of tensor graphs, and algorithmic improvements based on exploiting a type of knowledge called unknown functional dependencies. The used tensor graphs capitalize on a highly optimized tensor operation (matrix multiplication) which brings orders of magnitude speedups in circuit training and evaluation. The exploitation of unknown functional dependencies yields exponential reductions in the size of tractable circuits and gives rise to the notion of causal treewidth for offering a corresponding complexity bound. Beyond computational efficiency, we discuss empirical evidence showing the promise of learning from a combination of data and knowledge, in terms of data hungriness and robustness against noise perturbations. Sometimes, however, an accurate Bayesian network structure may not be available due to the incompleteness of human knowledge, leading to modeling errors in the form of missing dependencies or missing variable values. On this front, we discuss another set of advances for recovering from certain types of modeling errors. This is achieved using Testing Bayesian networks which dynamically select parameters based on the input evidence, and come with theoretical guarantees on full recovery under certain conditions.

  PublisherDOI

• Knowledge compilation

  Annals of Mathematics and Artificial Intelligence · 2024-05-17 · 3 citations

  articleOpen access1st authorCorresponding
  PublisherOA PDFDOI

• A New Class of Explanations for Classifiers with Non-Binary Features

  arXiv (Cornell University) · 2023-04-28 · 2 citations

  preprintOpen accessSenior author

  Two types of explanations have been receiving increased attention in the literature when analyzing the decisions made by classifiers. The first type explains why a decision was made and is known as a sufficient reason for the decision, also an abductive explanation or a PI-explanation. The second type explains why some other decision was not made and is known as a necessary reason for the decision, also a contrastive or counterfactual explanation. These explanations were defined for classifiers with binary, discrete and, in some cases, continuous features. We show that these explanations can be significantly improved in the presence of non-binary features, leading to a new class of explanations that relay more information about decisions and the underlying classifiers. Necessary and sufficient reasons were also shown to be the prime implicates and implicants of the complete reason for a decision, which can be obtained using a quantification operator. We show that our improved notions of necessary and sufficient reasons are also prime implicates and implicants but for an improved notion of complete reason obtained by a new quantification operator that we also define and study.

  PublisherOA PDFDOI

Recent grants

• RI: Probabilistic Reasoning with Bounded Computational Resources

  NSF · $450k · 2007–2011

• RI: Medium: Sentential Decision Diagrams

  NSF · $706k · 2015–2019

• RI: Small: Reasoning About the Behavior of Artificial Intelligence Systems

  NSF · $500k · 2019–2023

• RI: Small: Generalized Anytime Probabilistic Inference

  NSF · $449k · 2011–2015

Frequent coauthors

• Arthur Choi

  Kennesaw State University

  84 shared

• Guy Van den Broeck

  University of California, Los Angeles

  25 shared

• Knot Pipatsrisawat

  18 shared

• Pierre Marquis

  Institut Universitaire de France

  16 shared

• Mark Chavira

  University of California, Los Angeles

  15 shared

• Hei Chan

  15 shared

• Isabelle Dufour

  Université de Sherbrooke

  11 shared

• Khaled S. Refaat

  10 shared

Education

• Ph.D., Computer Science

  University of California, Los Angeles

  1990

• M.S., Computer Science

  University of California, Los Angeles

  1986

• B.S., Computer Science

  American University of Beirut

  1983

Awards & honors

• Editor-in-Chief, the Journal of Artificial Intelligence Rese…
• AAAI Fellow