Specializations
- Artificial Intelligence Ethics
- Machine Learning
- Natural Language Processing and Computer Vision
Research Areas
Biography
Aylin Caliskan is an assistant professor in the Information School at the University of Washington. Caliskan's research interests lie in artificial intelligence (AI) ethics, bias in AI, machine learning, and the implications of machine intelligence on equity. She investigates the reasoning behind biased AI representations and decisions by developing theoretically grounded statistical methods that uncover and quantify the biases of machines. Building these transparency enhancing algorithms involves the use of machine learning, natural language processing, and computer vision to interpret AI and gain insights about bias in machines as well as society. Caliskan's publication in Science demonstrated how semantics derived from language corpora contain human-like biases. Her work on machine learning's impact on fairness and privacy received the best talk and best paper awards. Caliskan was selected as a Rising Star in EECS at Stanford University. Caliskan holds a Ph.D. in Computer Science from Drexel University's College of Computing & Informatics and a Master of Science in Robotics from the University of Pennsylvania. Caliskan was a Postdoctoral Researcher and a Fellow at Princeton University's Center for Information Technology Policy. In 2021, Caliskan was named a Nonresident Fellow at the Brookings Institution.
Publications and Contributions
-
Conference PaperContrastive Visual Semantic Pretraining Magnifies the Semantics of Natural Language Representations (2022)60th Annual Meeting of the Association for Computational Linguistics (ACL), 2022
-
Conference PaperDetecting Emerging Associations and Behaviors With Regional and Diachronic Word Embeddings (2022)IEEE International Conference on Semantic Computing
-
Conference PaperDetecting Emerging Associations and Behaviors with Regional and Diachronic Word Embeddings (2022)2022 IEEE 16th International Conference on Semantic Computing (ICSC)
-
Conference PaperVAST: The Valence-Assessing Semantics Test for Contextualizing Language Models (2022)Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI 2022)
-
Conference PaperVAST: The Valence-Assessing Semantics Test for Contextualizing Language Models (2022)AAAI Conference on Artificial Intelligence
-
Journal Article, Academic JournalA Set of Maximally Distinct Facial Traits Learned by Machines is not Predictive of Appearance Bias in the Wild (2021)AI and Ethics
-
Conference PaperAutomatically Characterizing Targeted Information Operations Through Biases Present in Discourse on Twitter (2021)IEEE International Conference on Semantic Computing (ICSC)
-
Journal Article, Academic JournalBias in Natural Language Processing (2021)Brookings
-
ReportComments in Response to ‘A Proposal for Identifying and Managing Bias in Artificial Intelligence’ from the National Institute of Standards and Technology (2021)Draft NIST Special Publication 1270
-
Comments with American Psychological Association (APA) in response to the The National Science Foundation (NSF) and the White House Office of Science and Technology Policy (OSTP) Request for Information (RFI) on an Implementation Plan for a National Artificial Intelligence Research Resource (NAIRR) (2021)
-
Conference PaperDetecting Emergent Intersectional Biases: Contextualized Word Embeddings Contain a Distribution of Human-like Biases (2021)AAAI/ACM Artificial Intelligence, Ethics, and Society (AAAI/ACM AIES)
-
ReportDetecting and mitigating bias in natural language processing (2021)The Brookings Institution
-
Conference PaperDisparate Impact of Artificial Intelligence Bias in Ridehailing Economy’s Price Discrimination Algorithms (2021)AAAI/ACM Artificial Intelligence, Ethics, and Society (AAAI/ACM AIES)
-
Conference PaperImage Representations Learned With Unsupervised Pre-Training Contain Human-like Biases (2021)ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT)
-
Conference PaperLow Frequency Names Exhibit Bias and Overfitting in Contextualizing Language Models (2021)Empirical Methods in Natural Language Processing (EMNLP)
-
Book, Chapter in Non-Scholarly Book-NewSocial Biases in Word Embeddings and Their Relation to Human Cognition (2021)The Atlas of Language Analysis in Psychology Editors: Morteza Deghani, Ryan Boyd
-
Conference PaperValNorm Quantifies Semantics to Reveal Consistent Valence Biases Across Languages and Over Centuries (2021)Empirical Methods in Natural Language Processing (EMNLP)
-
Conference PaperIf I Tap It, Will They Come? An Introductory Analysis of Fairness in a Large-Scale Ride Hailing Dataset (2020)Academy of Marketing Science Annual Conference (AMS)
-
DocketComments in response to the National Institute of Standards and Technology Request for Information on Developing a Federal AI Standards Engagement Plan (2019)National Institute of Standards and Technology (NIST)
-
Conference PaperGit Blame Who?: Stylistic Authorship Attribution of Small, Incomplete Source Code Fragments (2019)Privacy Enhancing Technologies Symposium (PETS)
-
Conference PosterPrivacy and Security via Machine Learning and Natural Language Processing. (2018)Cybersecurity Retreat, Princeton University
-
Conference PaperStylistic authorship attribution of small, incomplete source code fragments Authors. (2018)IEEE/ACM 40th International Conference on Software Engineering: Companion
-
Journal Article, Academic JournalStylometry of Author-Specific and Country-Specific Style Features in JavaScript. (2018)Network and Distributed System Security Symposium (NDSS)
-
Conference PaperWhen Coding Style Survives Compilation: De-anonymizing Programmers from Executable Binaries (2018)Network and Distributed System Security Symposium (NDSS)
-
Journal Article, Academic JournalSemantics derived automatically from language corpora contain human-like biases (2017)Science
-
Conference PaperA Story of Discrimination and Unfairness (2016)Hot Topics in Privacy Enhancing Technologies (HotPETs)
-
Conference PaperDe-anonymizing Programmers via Code Stylometry (2015)USENIX Security Symposium (USENIX Security)
-
Conference PaperHow do we decide how much to reveal? (Hint: Our privacy behavior might be socially constructed.) (2015)Special Issue on Security, Privacy, and Human Behavior, ACM Computers & Society
-
Conference PaperDoppelgänger Finder: Taking Stylometry To The Underground (2014)IEEE Symposium on Security and Privacy
-
Conference PosterDoppelgänger Finder: Taking Stylometry To The Underground. (2014)Computer Science PhD Open House
-
Conference Workshop PaperPrivacy Detective: Detecting Private Information and Collective Privacy Behavior in a Large Social Network (2014)Workshop on Privacy in the Electronic Society (WPES)
-
Conference Workshop PaperApproaches to Adversarial Drift (2013)ACM Workshop on Artificial Intelligence and Security (AISec)
-
Conference Workshop PaperFrom Language to Family and Back: Native Language and Language Family Identification from English Text (2013)Conference of the North American Chapter of the Association for Computational Linguistics: Student Research Workshop (NAACL-SRW)
-
Conference Workshop PaperHow Privacy Flaws Affect Consumer Perception (2013)3rd Workshop on Socio-Technical Aspects in Security and Trust (STAST)
-
Conference PaperTranslate once, translate twice, translate thrice and attribute: Identifying authors and machine translation tools in translated text (2012)IEEE International Conference on Semantic Computing (ICSC)
-
Conference PaperUse Fewer Instances of the Letter “i”: Toward Writing Style Anonymization (2012)Privacy Enhancing Technologies Symposium (PETS)
-
Conference PosterENVOY: Exploration and Navigation Vehicle for geolOgY (2011)University of Pennsylvania’s Entry in NASA/NIA RASC-AL
-
Conference PaperLearning to Behave: Improving Covert Channel Security with Behavior-Based DesignThe 22nd Privacy Enhancing Technologies Symposium (PETS), 2022
Presentations
-
Artificial Intelligence for Social Good: When Machines Learn Human-like Biases from Data
(2021)
Harvard University, University of Chicago, University of Washington, Ninth Circuit’s Fairness Committee - Zoom
-
Workshop on AI on Artificial Intelligence in Information Research and Practice
(2021)
ASIS&T 2021 - Salt Lake City, Utah
-
Algorithmic Measures of Language Mirror Human Biases
(2020)
Georgetown University
-
Algorithmic Measures of Language Mirror Human Biases and Widely Shared Associations
(2020)
Santa Fe Institute
-
Bias and AI Ethics
(2020)
DefCon28 AI Village
-
Bias in AI
(2020)
NIST AI Workshop
-
Bias in AI and Digital Humanities
(2020)
University of Pennsylvania
-
Gender Breakthrough
(2020)
AI for Good Global Summit
-
Gender Equity
(2020)
AI for Good Global Summit
-
Implications of Biased AI on Democracy, Equity, and Justice
(2020)
COLING Workshop on Natural Language Processing for Internet Freedom
-
Promises and Pitfalls of Big Data Approaches to Intersectional Equity in STEM
(2020)
NSF Workshop
-
AI for Social Good, Bias and Ethics Panel
(2019)
WeCNLP Summit at Facebook
-
Algorithmic Measures of Language Mirror Human Biases
(2019)
Symposium on Computer-Resident Language and Naturalistic Conversation as Windows Into Social Cognition
-
Algorithmic Mirrors of Human Biases
(2019)
Virginia Tech
-
Algorithmic Mirrors of Human Biases
(2019)
University of Chicago
-
Algorithmic Mirrors of Society
(2019)
University of Maryland
-
Bias in AI
(2019)
Social Science Foo Camp at Facebook
-
Hands-on Tutorial: AI Fairness 360
(2019)
ACM Conference on Fairness, Accountability, and Transparency (ACM FAT*)
-
Human-like Bias in Machine Intelligence
(2019)
George Washington University, SEH WOW Talk Series
-
Monitoring Hate Speech in the US Media
(2019)
Workshop on Defining, Monitoring and Countering Hate Speech. George Washington University, School of Media and Public Affairs
-
Neural Networks for NLP
(2019)
George Washington University
-
NSF Workshop: Fairness, Ethics, Accountability, and Transparency (FEAT)
(2019)
NSF Workshop
-
Tutorial on Distributional Semantics via Word Embeddings
(2019)
Department of Psychology, Harvard University
-
AI & Equity
(2018)
MIT Media Lab
-
Bias in Machine Learning
(2018)
ACM & Women in Computer Science at GWU
-
De-anonymizing Programmers from Source Code and Binaries
(2018)
DEFCON
-
The Great Power of AI: Algorithmic Mirrors of Individuals and Society
(2018)
Brown University, Duke University, ETH Zurich, George Washington University, Tufts University, University of Maryland, University of Virginia, and Yale University
-
The Great Power of AI: Algorithmic Mirrors of Society
(2018)
DEFCON
-
Beyond Big Data: What Can We Learn from AI Models?
(2017)
AISec - CCS Workshop
-
A Story of Discrimination and Unfairness: Implicit Bias Embedded in Language Models
(2016)
HotPETS 2016 - PETS
-
De-anonymizing Programmers and Code Stylometry - Large Scale Authorship Attribution from Source Code and Executable Binaries of Compiled Code
(2016)
Princeton University CITP Luncheon Speaker Series
-
Natural Language Processing and Privacy: A Double Edged Sword
(2016)
Infer - PETS Workshop
-
Code Stylometry and Programmer De-anonymization
(2015)
Go¨ttingen University
-
De-anonymizing Programmers
(2015)
32C3 - Chaos Communication Congress
-
De-anonymizing Programmers via Code Stylometry
(2015)
Cornell Systems Lunch
-
Support Vector Machines, Kernel Methods, Random Forests, and Feature Projection
(2015)
CS613-Machine Learning
-
Security Review of Digital Privacy and the Underground: Miscreant Activity in the Internet Guest Lecture
(2014)
CS475-Computer and Network Security
-
Source Code and Cross-Domain Stylometry
(2014)
31st Chaos Communication Congress
-
Stylometry and Online Underground Markets
(2012)
29th Chaos Communication Congress
-
Quantifying the Translator Effect: Identifying authors and machine translation tools in translated text
(2011)
Girl Geek Dinners Philly