paper-with-me

홈 › Papers

Online Algorithm for Aggregating Experts' Predictions with Unbounded Quadratic Loss

2025-01-11 · Alexander Korotin, Vladimir V'yugin, Evgeny Burnaev

We consider the problem of online aggregation of expert predictions with the quadratic loss function. We propose an algorithm for aggregating expert predictions which does not require a prior knowledge of the upper bound on the losses. The algorithm is based on the exponential reweighing of expert losses.

📄 PDF Abstract BibTeX arXiv:2501.06505

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Online Aggregation of Unbounded Losses Using Shifting Experts with Confidence

2018-08-02 · Vladimir V'yugin, Vladimir Trunov

We develop the setting of sequential prediction based on shifting experts and on a "smooth" version of the method of specialized experts. To aggregate experts predictions, we use the AdaHedge algorithm, which is a versio…

valid

No-Regret Online Prediction with Strategic Experts

2023-05-24 · NeurIPS 2023 11

We study a generalization of the online binary prediction with expert advice framework where at each round, the learner is allowed to pick $m\geq 1$ experts from a pool of $K$ experts and the overall utility is a modular…

Prediction

Online Aggregation of Probability Forecasts with Confidence

2021-09-29 · Vladimir V'yugin, Vladimir Trunov

The paper presents numerical experiments and some theoretical developments in prediction with expert advice (PEA). One experiment deals with predicting electricity consumption depending on temperature and uses real data.…

scoring rule

Soft-Bayes: Prod for Mixtures of Experts with Log-Loss

2019-01-08 · Laurent Orseau, Tor Lattimore, Shane Legg

We consider prediction with expert advice under the log-loss with the goal of deriving efficient and robust algorithms. We argue that existing algorithms such as exponentiated gradient, online gradient descent and online…

Gaussian Experts Selection using Graphical Models

2021-02-02 · Hamed Jalali, Martin Pawelczyk, Gjergji Kasneci

Local approximations are popular methods to scale Gaussian processes (GPs) to big data. Local approximations reduce time complexity by dividing the original dataset into subsets and training a local expert on each subset…

Gaussian ProcessesUncertainty Quantification