296 citations · 575 across the 7 of their papers we have counts for
5 papers · 1 filter
How to post-train on a surrogate: Envelope sampling mitigates reward hacking
Sanjit Dandapanthula, Shuvom Sadhuka, Samir Khan +3
Large language models (LLMs) are commonly post-trained against LLM judges and other cheap surrogates because the true reward, such as human preference, is too expensive to query at…
Doubting AI Predictions: Influence-Driven Second Opinion Recommendation
Maria De-Arteaga, Alexandra Chouldechova, Artur Dubrawski
Effective human-AI collaboration requires a system design that provides humans with meaningful ways to make sense of and critically evaluate algorithmic recommendations. In this pa…
Leveraging Expert Consistency to Improve Algorithmic Decision Support
Maria De-Arteaga, Vincent Jeanselme, Artur Dubrawski +1
Machine learning (ML) is increasingly being used to support high-stakes decisions. However, there is frequently a construct gap: a gap between the construct of interest to the deci…
What's in a Name? Reducing Bias in Bios without Access to Protected Attributes
Alexey Romanov, Maria De-Arteaga, Hanna Wallach +7
There is a growing body of work that proposes methods for mitigating bias in machine learning systems. These methods typically rely on access to protected attributes such as race,…
Learning under selective labels in the presence of expert consistency
Maria De-Arteaga, Artur Dubrawski, Alexandra Chouldechova
We explore the problem of learning under selective labels in the context of algorithm-assisted decision making. Selective labels is a pervasive selection bias problem that arises w…