Zobrazeno 1 - 10
of 1 045
pro vyhledávání: '"Ustimenko, A."'
Online controlled experiments, colloquially known as A/B-tests, are the bread and butter of real-world recommender system evaluation. Typically, end-users are randomly assigned some system variant, and a plethora of metrics are then tracked, collecte
Externí odkaz:
http://arxiv.org/abs/2407.20665
Autor:
Jeunen, Olivier, Ustimenko, Aleksei
The off-policy paradigm casts recommendation as a counterfactual decision-making task, allowing practitioners to unbiasedly estimate online metrics using offline data. This leads to effective evaluation metrics, as well as learning procedures that di
Externí odkaz:
http://arxiv.org/abs/2405.10024
Autor:
Jeunen, Olivier, Mandav, Jatin, Potapov, Ivan, Agarwal, Nakul, Vaid, Sourabh, Shi, Wenzhe, Ustimenko, Aleksei
Real-world recommender systems often need to balance multiple objectives when deciding which recommendations to present to users. These include behavioural signals (e.g. clicks, shares, dwell time), as well as broader objectives (e.g. diversity, fair
Externí odkaz:
http://arxiv.org/abs/2405.02141
Autor:
Jeunen, Olivier, Ustimenko, Aleksei
Online controlled experiments are a crucial tool to allow for confident decision-making in technology companies. A North Star metric is defined (such as long-term revenue or user retention), and system variants that statistically significantly improv
Externí odkaz:
http://arxiv.org/abs/2402.03915
Online controlled experiments, such as A/B-tests, are commonly used by modern tech companies to enable continuous system improvements. Despite their paramount importance, A/B-tests are expensive: by their very definition, a percentage of traffic is a
Externí odkaz:
http://arxiv.org/abs/2401.04062
Many platforms on the web present ranked lists of content to users, typically optimized for engagement-, satisfaction- or retention- driven metrics. Advances in the Learning-to-Rank (LTR) research literature have enabled rapid growth in this applicat
Externí odkaz:
http://arxiv.org/abs/2401.04053
Autor:
Jeunen, Olivier, Sagtani, Hitesh, Doi, Himanshu, Karimov, Rasul, Pokharna, Neeti, Kalim, Danish, Ustimenko, Aleksei, Green, Christopher, Shi, Wenzhe, Mehrotra, Rishabh
Practitioners who wish to build real-world applications that rely on ranking models, need to decide which modelling paradigm to follow. This is not an easy choice to make, as the research literature on this topic has been shifting in recent years. In
Externí odkaz:
http://arxiv.org/abs/2312.01760
Autor:
Volkov, Ilya, Ustimenko, Nikita, Kornovan, Danil, Savelev, Roman, Sheremet, Alexandra, Petrov, Mihail
The optically trapped ensembles of atoms provide a versatile platform for storing and coherent manipulation of quantum information. However, efficient realization of quantum information processing requires long-lived quantum states protected from the
Externí odkaz:
http://arxiv.org/abs/2310.06791
In this work, we consider rather general and broad class of Markov chains, Ito chains, that look like Euler-Maryama discretization of some Stochastic Differential Equation. The chain we study is a unified framework for theoretical analysis. It comes
Externí odkaz:
http://arxiv.org/abs/2310.06081
Autor:
Ustimenko, Nikita, Kornovan, Danil, Volkov, Ilya, Sheremet, Alexandra, Savelev, Roman, Petrov, Mihail
Arrays of coupled dipole emitters support collective single- and multiphoton states that can preserve quantum excitations. One of the crucial characteristics of these states is the lifetime, which is fundamentally limited due to spontaneous emission.
Externí odkaz:
http://arxiv.org/abs/2309.14461