Dana Angluin

Dana Angluin is an American computer scientist noted for foundational work in computational learning theory, algorithms for distributed computing, and the formalization of learning models such as the query model and the equivalence query framework. Her work influenced research in machine learning, cryptography, complexity theory, and automata theory. Angluin's models and algorithms are taught in courses at institutions including Stanford University, Massachusetts Institute of Technology, Harvard University, and Princeton University.

Early life and education

Angluin was born in Newark, New Jersey and raised in a period when computing emerged alongside institutions like Bell Labs, IBM and RAND Corporation. She completed undergraduate studies at Princeton University during an era shaped by figures such as John Nash and Alan Turing's legacy, then earned a Ph.D. at Stanford University under the supervision of Robert W. Floyd, joining a lineage that includes Donald Knuth and John McCarthy. Her doctoral work intersected with themes explored at Carnegie Mellon University, University of California, Berkeley, and during collaborations connected to DARPA research initiatives.

Academic career and positions

Angluin held a long-term faculty position at Yale University, where she worked alongside faculty from departments such as Computer Science Department, Yale University and engaged with scholars from Cornell University, Columbia University, and University of Pennsylvania. She visited research centers and universities including Massachusetts Institute of Technology, University of California, Berkeley, Harvard University, and international institutions connected to École Polytechnique Fédérale de Lausanne and University of Oxford. Her service includes participation in program committees for conferences like STOC, FOCS, COLT, ICML, and SODA, and collaboration with researchers affiliated with Microsoft Research, Google Research, and Bell Labs.

Research contributions and theories

Angluin introduced and formalized influential models in computational learning theory, notably the exact learning model with membership and equivalence queries, which shaped subsequent work by researchers at MIT, Stanford University, and UC Berkeley. Her seminal algorithms for learning regular languages from queries connected to the theory of finite automata and influenced studies in formal languages and pattern matching at places like Bell Labs and IBM Research. She made key contributions to distributed verification and fault tolerance, intersecting with research on consensus protocols such as Paxos and Raft, and related to impossibility results like the FLP impossibility result. Angluin's work also impacted cryptography via insights on query models that informed security analyses at RSA Laboratories and influenced protocol verification efforts at NIST and IETF.

Her publications advanced connections among computational complexity theory, property testing, and probabilistic algorithms, informing research at venues including SIAM, ACM, and IEEE. Collaborations and citations span scholars such as Leslie Valiant, Michael Sipser, Dana Scott, Eli Upfal, Ravi Kannan, and Christos Papadimitriou. Through her models, subsequent research on active learning at Google Research, semi-supervised learning at Facebook AI Research, and algorithmic learning at DeepMind drew theoretical foundations from her results.

Selected publications

- "Queries and Concept Learning" (seminal paper developing membership and equivalence queries), cited in work from Leslie Valiant, Michael Kearns, and Avrim Blum; influential at COLT and ICML. - Papers on learning regular sets and automata, connecting to the Angluin algorithm used in subsequent studies at MIT and UC Berkeley. - Articles on distributed verification and fault tolerance influencing research at Stanford University and Harvard University's distributed systems groups; frequently cited alongside work from Leslie Lamport and Nancy Lynch. - Contributions to computational models bridging probabilistic computation and learnability, referenced in contexts including PP and BPP complexity studies, and in textbooks by Michael Sipser and Martha Pollack. - Survey chapters and invited papers in volumes associated with SIAM and ACM collections, used in curricula at Princeton University and Carnegie Mellon University.

Awards and honors

Angluin's work has been recognized through citations and invited talks at major conferences such as STOC, FOCS, COLT, and ICML. Her papers are frequently listed among influential works in retrospective collections by ACM and IEEE. She has held visiting scholar positions and received grants and fellowships from agencies including NSF, DARPA, and foundations associated with MacArthur Foundation-era support for computing research. Angluin's contributions are celebrated in historical overviews of computational learning theory and in festschrifts honoring pioneers like Leslie Valiant and Dana Scott.

Category:American computer scientists Category:Women in computer science Category:Machine learning researchers