Cynthia Dwork
In plain wordsAn American computer scientist who co-invented differential privacy, the leading mathematical definition of privacy in data analysis, and helped found the study of fairness in algorithms.
American theoretical computer scientist at Harvard University, previously at IBM Research and Microsoft Research, whose definitions shaped privacy and fairness in machine learning. With Frank McSherry, Kobbi Nissim and Adam Smith she introduced differential privacydifferential privacy, n.: A mathematical promise that the results of analyzing a dataset — or a model trained on it — would look almost the same whether or not any one person's data was included, so little can be learned about that person. (2006), which guarantees that an analysis reveals almost nothing about any single individual by adding calibrated noise; it underlies DP-SGDDP-SGD, abbr.: A way of training a neural network so that it cannot learn too much about any single person in the data: each example's influence is capped, and random noise is added at every step. training and the US Census Bureau's 2020 data products, and won the four the 2017 Gödel Prize. Her 'Fairness Through Awareness' (2012, with Moritz Hardt, Toniann Pitassi, Omer Reingold and Richard Zemel) defined individual fairnessindividual fairness, n.: The idea that a fair system should treat similar people similarly: two applicants who are alike in every way that matters should get similar outcomes., the principle that similar individuals should be treated similarly, and exposed the weaknesses of statistical paritydemographic parity, n.: A fairness test that asks whether a system says yes (approves the loan, shortlists the applicant) at the same rate for every group, whatever other differences exist between them., helping found the field of algorithmic fairnessalgorithmic fairness, n.: The effort to make automated decisions fair: to say in numbers what fair treatment means, measure whether a system achieves it, and correct the system if it does not.. Earlier she made foundational contributions to distributed consensus and cryptography, including non-malleable cryptography and, with Moni Naor, the 'pricing via processing' idea (1992) that anticipated proof-of-work. Honors include the Knuth Prize (2020) and the US National Medal of Science (2025).
Related
- differential privacydifferential privacy, n.: A mathematical promise that the results of analyzing a dataset — or a model trained on it — would look almost the same whether or not any one person's data was included, so little can be learned about that person.
- DP-SGDDP-SGD, abbr.: A way of training a neural network so that it cannot learn too much about any single person in the data: each example's influence is capped, and random noise is added at every step.
- individual fairnessindividual fairness, n.: The idea that a fair system should treat similar people similarly: two applicants who are alike in every way that matters should get similar outcomes.
- algorithmic fairnessalgorithmic fairness, n.: The effort to make automated decisions fair: to say in numbers what fair treatment means, measure whether a system achieves it, and correct the system if it does not.
- demographic paritydemographic parity, n.: A fairness test that asks whether a system says yes (approves the loan, shortlists the applicant) at the same rate for every group, whatever other differences exist between them.
- membership inference attackmembership inference attack, n.: An attack that tries to tell whether a particular person's record, or a particular document, was part of an AI model's training data — which can itself reveal private facts.
Sources
- Dwork, C., McSherry, F., Nissim, K. & Smith, A. (2006). Calibrating Noise to Sensitivity in Private Data Analysis. TCC 2006.
- Dwork, C., Hardt, M., Pitassi, T., Reingold, O. & Zemel, R. (2012). Fairness Through Awareness. ITCS 2012.