Zobrazeno 1 - 10
of 327
pro vyhledávání: '"Kallus, Nathan"'
In multi-armed bandits, the tasks of reward maximization and pure exploration are often at odds with each other. The former focuses on exploiting arms with the highest means, while the latter may require constant exploration across all arms. In this
Externí odkaz:
http://arxiv.org/abs/2410.15564
Autor:
Lindon, Michael, Kallus, Nathan
Motivated by monitoring the arrival of incoming adverse events such as customer support calls or crash reports from users exposed to an experimental product change, we consider sequential hypothesis testing of continuous-time inhomogeneous Poisson po
Externí odkaz:
http://arxiv.org/abs/2410.09282
Conformal Prediction methods have finite-sample distribution-free marginal coverage guarantees. However, they generally do not offer conditional coverage guarantees, which can be important for high-stakes decisions. In this paper, we propose a novel
Externí odkaz:
http://arxiv.org/abs/2409.17466
This paper illustrates the central role of loss functions in data-driven decision making, providing a comprehensive survey on their influence in cost-sensitive classification (CSC) and reinforcement learning (RL). We demonstrate how different regress
Externí odkaz:
http://arxiv.org/abs/2409.12799
Autor:
Cho, Brian M, Pop, Ana-Roxana, Gan, Kyra, Corbett-Davies, Sam, Nir, Israel, Evnine, Ariel, Kallus, Nathan
When modifying existing policies in high-risk settings, it is often necessary to ensure with high certainty that the newly proposed policy improves upon a baseline, such as the status quo. In this work, we consider the problem of safe policy improvem
Externí odkaz:
http://arxiv.org/abs/2408.12004
Jackknife instrumental variable estimation (JIVE) is a classic method to leverage many weak instrumental variables (IVs) to estimate linear structural models, overcoming the bias of standard methods like two-stage least squares. In this paper, we ext
Externí odkaz:
http://arxiv.org/abs/2406.14140
Autor:
Oprescu, Miruna, Kallus, Nathan
Accurately predicting conditional average treatment effects (CATEs) is crucial in personalized medicine and digital platform analytics. Since the treatments of interest often cannot be directly randomized, observational data is leveraged to learn CAT
Externí odkaz:
http://arxiv.org/abs/2406.06452
Contextual linear optimization (CLO) uses predictive contextual features to reduce uncertainty in random cost coefficients and thereby improve average-cost performance. An example is the stochastic shortest path problem with random edge costs (e.g.,
Externí odkaz:
http://arxiv.org/abs/2405.16564
Autor:
He, Zhankui, Xie, Zhouhang, Steck, Harald, Liang, Dawen, Jha, Rahul, Kallus, Nathan, McAuley, Julian
Large language models (LLMs) are revolutionizing conversational recommender systems by adeptly indexing item content, understanding complex conversational contexts, and generating relevant item titles. However, controlling the distribution of recomme
Externí odkaz:
http://arxiv.org/abs/2405.12119
Autor:
Bibaut, Aurélien, Kallus, Nathan
Adaptive experiments such as multi-arm bandits adapt the treatment-allocation policy and/or the decision to stop the experiment to the data observed so far. This has the potential to improve outcomes for study participants within the experiment, to i
Externí odkaz:
http://arxiv.org/abs/2405.01281