• Publications
  • Influence
Big Data's Disparate Impact
Advocates of algorithmic techniques like data mining argue that these techniques eliminate human biases from the decision-making process. But an algorithm is only as good as the data it works with.Expand
  • 1,070
  • 53
  • PDF
Fairness and Abstraction in Sociotechnical Systems
TLDR
A key goal of the fair-ML community is to develop machine-learning based systems that, once introduced into a social context, can achieve social and legal outcomes such as fairness, justice, and due process. Expand
  • 193
  • 20
  • PDF
The hidden assumptions behind counterfactual explanations and principal reasons
TLDR
We demonstrate that the utility of feature-highlighting explanations relies on a number of easily overlooked assumptions: that the recommended change in feature values clearly maps to real-world actions, that features can be made commensurate by looking only at the distribution of the training data, and that features are only relevant to the decision at hand. Expand
  • 40
  • 7
  • PDF
"Meaningful Information" and the Right to Explanation
TLDR
There is no single, neat statutory provision labeled the “right to explanation” in Europe’s new General Data Protection Regulation (GDPR). Expand
  • 155
  • 6
The Intuitive Appeal of Explainable Machines
TLDR
We show that machine learning models can be both inscrutable and nonintuitive and that these are related, but distinct, properties. Expand
  • 99
  • 5
  • PDF
Disparate Impact in Big Data Policing
Police departments large and small have begun to use data mining techniques to predict the where, when, and who of crime before it occurs. But data mining systems can have a disproportionatelyExpand
  • 63
  • 2
  • PDF
Contextual Expectations of Privacy
Fourth Amendment search jurisprudence is nominally based on a “reasonable expectation of privacy,” but actual doctrine is disconnected from society’s conception of privacy. Courts rely on variousExpand
  • 10
  • 2
  • PDF
Negligence and AI's Human Users
TLDR
This Article argues that AI poses serious challenges for negligence law, the governing regime when harm comes from users of AI. Expand
  • 5
  • PDF
Clock Division as a Power Saving Strategy in a System Constrained by High Transmission Frequency and Low Data Rate
Systems are often restricted to have higher transmission frequency than required by their data rates. Possible constraints include channel attenuation, power requirements, and backward compatibility.Expand
  • 3
...
1
2
...