no code implementations • 27 May 2023 • Smitha Milli, Emma Pierson, Nikhil Garg
Many recommender systems are based on optimizing a linear weighting of different user behaviors, such as clicks, likes, shares, etc.
no code implementations • 19 Jul 2021 • Smitha Milli, Luca Belli, Moritz Hardt
Our results suggest that observational studies derived from user self-selection are a poor alternative to randomized experimentation on online platforms.
no code implementations • 21 Aug 2020 • Smitha Milli, Luca Belli, Moritz Hardt
Most recommendation engines today are based on predicting user engagement, e. g. predicting whether a user will click on an item or not.
no code implementations • NeurIPS 2020 • Hong Jun Jeon, Smitha Milli, Anca D. Dragan
It is often difficult to hand-specify what the correct reward function is for a task, so researchers have instead aimed to learn reward functions from human behavior or feedback.
no code implementations • 3 Dec 2019 • Ravit Dotan, Smitha Milli
As machine learning models are increasingly used for high-stakes decision making, scholars have sought to intervene to ensure that such models do not encode undesirable social and political values.
no code implementations • ICML 2020 • John Miller, Smitha Milli, Moritz Hardt
Moreover, we show a similar result holds for designing cost functions that satisfy the requirements of previous work.
no code implementations • 9 Mar 2019 • Smitha Milli, Anca D. Dragan
In this work, we focus on misspecification: we argue that robots might not know whether people are being pedagogic or literal and that it is important to ask which assumption is safer to make.
no code implementations • 25 Aug 2018 • Smitha Milli, John Miller, Anca D. Dragan, Moritz Hardt
Consequential decision-making typically incentivizes individuals to behave strategically, tailoring their behavior to the specifics of the decision rule.
no code implementations • 13 Jul 2018 • Smitha Milli, Ludwig Schmidt, Anca D. Dragan, Moritz Hardt
We show through theory and experiment that gradient-based explanations of a model quickly reveal the model itself.
1 code implementation • NeurIPS 2017 • Dylan Hadfield-Menell, Smitha Milli, Pieter Abbeel, Stuart Russell, Anca Dragan
When designing the reward, we might think of some specific training scenarios, and make sure that the reward will lead to the right behavior in those scenarios.
no code implementations • ICLR 2018 • Smitha Milli, Pieter Abbeel, Igor Mordatch
Teachers intentionally pick the most informative examples to show their students.
1 code implementation • 28 May 2017 • Smitha Milli, Dylan Hadfield-Menell, Anca Dragan, Stuart Russell
We show that when a human is not perfectly rational then a robot that tries to infer and act according to the human's underlying preferences can always perform better than a robot that simply follows the human's literal order.