Следене
Aleksei Ustimenko
Aleksei Ustimenko
Staff Applied Scientist at ShareChat. Ex-Yandex.Research
Потвърден имейл адрес: sharechat.co
Заглавие
Позовавания
Позовавания
Година
Uncertainty in Gradient Boosting via Ensembles
A Malinin, L Prokhorenkova, A Ustimenko
International Conference on Learning Representations, 2021, 2020
982020
SGLB: Stochastic Gradient Langevin Boosting
A Ustimenko, L Prokhorenkova
International Conference on Machine Learning, 2021, 2020
27*2020
StochasticRank: Global Optimization of Scale-Free Discrete Functions
A Ustimenko, L Prokhorenkova
International Conference on Machine Learning, 2020, 2020
172020
On (Normalised) Discounted Cumulative Gain as an Offline Evaluation Metric for Top-𝑛 Recommendation
O Jeunen, I Potapov, A Ustimenko
arXiv preprint arXiv:2307.15053, 2023
72023
Which Tricks are Important for Learning to Rank?
I Lyzhin, A Ustimenko, A Gulin, L Prokhorenkova
International Conference on Machine Learning, 2023, 2022
72022
Learning to select for a predefined ranking
A Ustimenko, A Vorobev, G Gusev, P Serdyukov
International Conference on Machine Learning, 6477-6486, 2019
72019
Method and server for training machine learning algorithm for ranking objects
AI Ustimenko
US Patent App. 17/223,680, 2021
42021
Variance Reduction in Ratio Metrics for Efficient Online Experiments
S Baweja, N Pokharna, A Ustimenko, O Jeunen
European Conference on Information Retrieval, 292-297, 2024
32024
On gradient boosted decision trees and neural rankers: a case-study on short-video recommendations at ShareChat
O Jeunen, H Sagtani, H Doi, R Karimov, N Pokharna, D Kalim, ...
Proceedings of the 15th Annual Meeting of the Forum for Information …, 2023
22023
Multi-Objective Recommendation via Multivariate Policy Learning
O Jeunen, J Mandav, I Potapov, N Agarwal, S Vaid, W Shi, A Ustimenko
arXiv preprint arXiv:2405.02141, 2024
12024
Learning-to-Rank with Nested Feedback
H Sagtani, O Jeunen, A Ustimenko
European Conference on Information Retrieval, 306-315, 2024
12024
Learning Metrics that Maximise Power for Accelerated A/B-Tests
O Jeunen, A Ustimenko
arXiv preprint arXiv:2402.03915, 2024
12024
Methods and systems for generating an uncertainty score for an output of a Gadient Boosting Decision Tree model
LA Prokhorenkova, AI Ustimenko, AA Malinin
US Patent App. 17/351,719, 2022
12022
Gradient Boosting Performs Gaussian Process Inference
A Ustimenko, A Beliakov, L Prokhorenkova
International Conference on Learning Representations, 2023, 2022
12022
Methods and systems for training a decision-tree based machine learning algorithm (mla)
AI Ustimenko
US Patent App. 17/207,403, 2022
12022
The triviality condition for kernels of quadratic mappings and its application in optimization methods
A Tret’yakov, E Tyrtyshnikov, A Ustimenko
Russian Journal of Numerical Analysis and Mathematical Modelling 32 (4), 267-274, 2017
12017
Methods and systems for training a decision-tree based machine learning algorithm (mla)
L Prokhorenkova, A Ustimenko
US Patent App. 18/395,024, 2024
2024
: Off-Policy Estimation with Pairs of Policies
O Jeunen, A Ustimenko
arXiv preprint arXiv:2405.10024, 2024
2024
Ito Diffusion Approximation of Universal Ito Chains for Sampling, Optimization and Boosting
A Ustimenko, A Beznosikov
International Conference on Learning Representations, 2024, 2023
2023
Deep Stochastic Mechanics
E Orlova, A Ustimenko, R Jiang, PY Lu, R Willett
arXiv preprint arXiv:2305.19685, 2023
2023
Системата не може да изпълни операцията сега. Опитайте отново по-късно.
Статии 1–20