Characterizing manipulation from AI systems
Manipulation is a concern in many domains, such as social media, advertising, and
chatbots. As AI systems mediate more of our digital interactions, it is important to understand …
chatbots. As AI systems mediate more of our digital interactions, it is important to understand …
Performative power
M Hardt, M Jagadeesan… - Advances in Neural …, 2022 - proceedings.neurips.cc
We introduce the notion of performative power, which measures the ability of a firm
operating an algorithmic system, such as a digital content recommendation platform, to …
operating an algorithmic system, such as a digital content recommendation platform, to …
Anticipating performativity by predicting from predictions
C Mendler-Dünner, F Ding… - Advances in neural …, 2022 - proceedings.neurips.cc
Predictions about people, such as their expected educational achievement or their credit
risk, can be performative and shape the outcome that they are designed to predict …
risk, can be performative and shape the outcome that they are designed to predict …
Supply-side equilibria in recommender systems
M Jagadeesan, N Garg… - Advances in Neural …, 2024 - proceedings.neurips.cc
Algorithmic recommender systems such as Spotify and Netflix affect not only consumer
behavior but also producer incentives. Producers seek to create content that will be shown …
behavior but also producer incentives. Producers seek to create content that will be shown …
Who leads and who follows in strategic classification?
As predictive models are deployed into the real world, they must increasingly contend with
strategic behavior. A growing body of work on strategic classification treats this problem as a …
strategic behavior. A growing body of work on strategic classification treats this problem as a …
Modeling content creator incentives on algorithm-curated platforms
Content creators compete for user attention. Their reach crucially depends on algorithmic
choices made by developers on online platforms. To maximize exposure, many creators …
choices made by developers on online platforms. To maximize exposure, many creators …
Regret minimization with performative feedback
M Jagadeesan, T Zrnic… - … on Machine Learning, 2022 - proceedings.mlr.press
In performative prediction, the deployment of a predictive model triggers a shift in the data
distribution. As these shifts are typically unknown ahead of time, the learner needs to deploy …
distribution. As these shifts are typically unknown ahead of time, the learner needs to deploy …
Information discrepancy in strategic learning
We initiate the study of the effects of non-transparency in decision rules on individuals' ability
to improve in strategic learning settings. Inspired by real-life settings, such as loan approvals …
to improve in strategic learning settings. Inspired by real-life settings, such as loan approvals …
Generalized strategic classification and the case of aligned incentives
S Levanon, N Rosenfeld - International Conference on …, 2022 - proceedings.mlr.press
Strategic classification studies learning in settings where self-interested users can
strategically modify their features to obtain favorable predictive outcomes. A key working …
strategically modify their features to obtain favorable predictive outcomes. A key working …
Performative prediction: Past and future
M Hardt, C Mendler-Dünner - arXiv preprint arXiv:2310.16608, 2023 - arxiv.org
Predictions in the social world generally influence the target of prediction, a phenomenon
known as performativity. Self-fulfilling and self-negating predictions are examples of …
known as performativity. Self-fulfilling and self-negating predictions are examples of …