Timur Garipov
Новосибирский государственный университет
- Adversarial Robustness in Machine Learning
- Domain Adaptation and Few-Shot Learning
- Natural Language Processing Techniques
- Advanced Neural Network Applications
- Gaussian Processes and Bayesian Inference
Чем занимается
Ключевые темы по публикациям: Adversarial Robustness in Machine Learning; Domain Adaptation and Few-Shot Learning; Natural Language Processing Techniques; Advanced Neural Network Applications; Gaussian Processes and Bayesian Inference.
Последние работы
Все 28 в OpenAlexAveraging Weights Leads to Wider Optima and Better Generalization2018 · аннотациясвернуть
Deep neural networks are typically trained by optimizing a loss function with an SGD variant, in conjunction with a decaying learning rate, until convergence. We show that simple averaging of multiple points along the…
Averaging Weights Leads to Wider Optima and Better Generalization2018 · аннотациясвернуть
Deep neural networks are typically trained by optimizing a loss function with an SGD variant, in conjunction with a decaying learning rate, until convergence. We show that simple averaging of multiple points along the…
Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNs2018 · аннотациясвернуть
The loss functions of deep neural networks are complex and their geometric properties are not well understood. We show that the optima of these complex loss functions are in fact connected by simple curves over which…
A Simple Baseline for Bayesian Uncertainty in Deep Learning2019 · аннотациясвернуть
We propose SWA-Gaussian (SWAG), a simple, scalable, and general purpose approach for uncertainty representation and calibration in deep learning. Stochastic Weight Averaging (SWA), which computes the first moment of…
Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNs2018 · аннотациясвернуть
The loss functions of deep neural networks are complex and their geometric properties are not well understood. We show that the optima of these complex loss functions are in fact connected by simple curves over which…
- OpenAI GPT-5 System Card2025
- OpenAI GPT-5 System Card2025
- An Experimental Study of Automating Explanatory Dictionary Compilation with Language Models2025
Наукометрия
Написать научруку
Email не найден в открытых источниках — поищи на странице вуза или в последних статьях.
Шаблон письма
Здравствуйте, Timur Garipov! Я студент(ка) [курс, факультет, вуз]. Мне интересна тема [опишите интересы своими словами]. Прочитал(а) вашу работу «Averaging Weights Leads to Wider Optima and Better Generalization» (2018) — она близка к тому, чем я хочу заниматься. Хочу обсудить возможность выполнить научную работу под вашим руководством. Буду благодарен(на) за ответ — готов(а) рассказать о себе подробнее и прислать резюме. С уважением, [Имя Фамилия]
На этапе теста письмо отправляешь сам из своей почты. Отправка из сервиса и статусы заявок — скоро.