Название исследуемой задачи:	Использование предикаторной функции для построения ансамбля нейросетей
Тип научной работы:	M1P
Автор:	Уденеев Александр Владимирович
Научный руководитель:	Бахтеев Олег
Научный консультант(при наличии):	Бабкин Петр

Abstract

The automated search for optimal neural network architectures (NAS) is a challenging compu- tational problem, and Neural Ensemble Search (NES) is even more complex. In this work, we propose a surrogate-based approach for ensebmle creation. Neural architectures are represented as graphs, and their predictions on a dataset serve as training data for the surrogate function. Using this function, we develop an efficient NES framework that enables the selection of diverse and high-performing architectures. The resulting ensemble achieves superior predictive accuracy on CIFAR-10 compared to other one-shot NES methods, demonstrating the effectiveness of our approach.

Keywords: NES, GCN, triplet loss, surrogate function

Research publications

Presentations at conferences on the topic of research

Software modules developed as part of the study

1. A python package mylib with all implementation here.
2. A code with all experiment visualisation here. Can use colab.