34 citations · 169 across the 29 of their papers we have counts for
34 papers
Knowing When to Intervene: Causal Evaluation of AI Agents
Ayush Sawarni, Wenqian Xing, Tijana Zrnic +2
Current evaluations of agentic AI systems primarily rely on benchmarks. However, the static nature of benchmarks makes them uninformative about the value of auditing interventions…
Reliable AI-Assisted Decisions via Selective Auditing
Isaac Gibbs, Elise L. Han, Ryan J. Tibshirani +1
We consider online monitoring of an AI system that produces a sequence of decisions. At each round, the system outputs a proposed decision together with a confidence score estimati…
High-Value Training Data from Noisy, Diverse Sources
Hedi Driss, Evan Frick, Anastasios N. Angelopoulos +4
Training state-of-the-art models increasingly requires learning from large, heterogeneous datasets whose examples vary in quality, domain relevance, and source of supervision. Data…
Partially Performative Prediction
Jaewook Lee, Tijana Zrnic
Performative prediction studies feedback loops that arise when predictive models are deployed in consequential domains. In these settings, deploying a model can change the populati…
Valid Inference with Synthetic Data via Task Exchangeability
Lezhi Tan, Tijana Zrnic
There is a proliferation of work arguing for the use of synthetic data in scientific research. For example, social scientists are arguing for the use of LLM-generated "silicon samp…
Look-Ahead Reasoning on Learning Platforms
Haiqing Zhu, Tijana Zrnic, Celestine Mendler-Dünner
On many learning platforms, the optimization criteria guiding model training reflect the priorities of the designer rather than those of the individuals they affect. Consequently,…
Robust Sampling for Active Statistical Inference
Puheng Li, Tijana Zrnic, Emmanuel Candès
Active statistical inference is a new method for inference with AI-assisted data collection. Given a budget on the number of labeled data points that can be collected and assuming…
Probably Approximately Correct Labels
Emmanuel J. Candès, Andrew Ilyas, Tijana Zrnic
Obtaining high-quality labeled datasets is often costly, requiring either human annotation or expensive experiments. In theory, powerful pre-trained AI models provide an opportunit…
Prediction-Powered Inference with Imputed Covariates and Nonuniform Sampling
Dan M. Kluger, Kerri Lu, Tijana Zrnic +2
Machine learning models are increasingly used to produce predictions that serve as input data in subsequent statistical analyses. For example, computer vision predictions of econom…
Predictions as Surrogates: Revisiting Surrogate Outcomes in the Age of AI
Wenlong Ji, Lihua Lei, Tijana Zrnic
We establish a formal connection between the decades-old surrogate outcome model in biostatistics and economics and the emerging field of prediction-powered inference (PPI). The co…
A Flexible Defense Against the Winner's Curse
Tijana Zrnic, William Fithian
Across science and policy, decision-makers often need to draw conclusions about the best candidate among competing alternatives. For instance, researchers may seek to infer the eff…
Can Unconfident LLM Annotations Be Used for Confident Conclusions?
Kristina Gligorić, Tijana Zrnic, Cinoo Lee +2
Large language models (LLMs) have shown high agreement with human raters across a variety of tasks, demonstrating potential to ease the challenges of human data collection. In comp…
A Note on the Prediction-Powered Bootstrap
Tijana Zrnic
We introduce PPBoot: a bootstrap-based method for prediction-powered inference. PPBoot is applicable to arbitrary estimation problems and is very simple to implement, essentially o…
Active Statistical Inference
Tijana Zrnic, Emmanuel J. Candès
Inspired by the concept of active learning, we propose active inference$\unicode{x2013}$a methodology for statistical inference with machine-learning-assisted data collection. Assu…
PPI++: Efficient Prediction-Powered Inference
Anastasios N. Angelopoulos, John C. Duchi, Tijana Zrnic
We present PPI++: a computationally lightweight methodology for estimation and inference based on a small labeled dataset and a typically much larger dataset of machine-learning pr…
Cross-Prediction-Powered Inference
Tijana Zrnic, Emmanuel J. Candès
While reliable data-driven decision-making hinges on high-quality labeled data, the acquisition of quality labels often involves laborious human annotations or slow and expensive s…
Plug-in Performative Optimization
Licong Lin, Tijana Zrnic
When predictions are performative, the choice of which predictor to deploy influences the distribution of future observations. The overarching goal in learning under performativity…
Algorithmic Collective Action in Machine Learning
Moritz Hardt, Eric Mazumdar, Celestine Mendler-Dünner +1
We initiate a principled study of algorithmic collective action on digital platforms that deploy machine learning algorithms. We propose a simple theoretical model of a collective…
Prediction-Powered Inference
Anastasios N. Angelopoulos, Stephen Bates, Clara Fannjiang +2
Prediction-powered inference is a framework for performing valid statistical inference when an experimental dataset is supplemented with predictions from a machine-learning system.…
Locally Simultaneous Inference
Tijana Zrnic, William Fithian
Selective inference is the problem of giving valid answers to statistical questions chosen in a data-driven manner. A standard solution to selective inference is simultaneous infer…
A Note on Zeroth-Order Optimization on the Simplex
Tijana Zrnic, Eric Mazumdar
We construct a zeroth-order gradient estimator for a smooth function defined on the probability simplex. The proposed estimator queries the simplex only. We prove that projected gr…
Valid Inference After Causal Discovery
Paula Gradu, Tijana Zrnic, Yixin Wang +1
Causal discovery and causal effect estimation are two fundamental tasks in causal inference. While many methods have been developed for each task individually, statistical challeng…
Regret Minimization with Performative Feedback
Meena Jagadeesan, Tijana Zrnic, Celestine Mendler-Dünner
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 t…
Who Leads and Who Follows in Strategic Classification?
Tijana Zrnic, Eric Mazumdar, S. Shankar Sastry +1
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 proble…