Best Paper Awards

Data Analysis and Machine Learning

Ensemble Clustering with Heterogeneous Transfer Learning
Vladimir Berikov

The author considers the important problem of clustering in presence of additional labeled dataset. He proposes to extract certain statistical features from the labelled dataset and then use them to refine clustering of the unlabelled data.

Computer Vision

DeepLOC: Deep Learning-based Bone Pathology Localization and Classification in Wrist X-ray Images.
Razan Dibo, Andrey Galichin, Pavel Astashev, Dmitry V. Dylov and Oleg Y. Rogov

The authors proposed a novel approach to detecting and classification of pathologies on wrist X-ray images. They proposed a new block in the transformer leading to the state-of-the-art performance in the considered problem.

Natural Language Processing

Benchmarking Multi-Label Topic Classification in Kyrgyz Language
Anton Alekseev, Sergey Nikolenko and Gulnara Kabaeva

For their meticulous work on creating NLP datasets for Kyrgyz: a low-resource language spoken by millions of people.

Social Network Analysis

Limit Distributions of Friendship Index in Scale-Free Networks
Sergei Sidorov, Sergei Mironov and Alexey Grigoriev

The paper explores the behavior of the friends index in scale-free graphs, with number of edges distributed as power law and treating it as a probabilistic model.

Theoretical Machine Learning and Optimization

Is Canfield Right? On the Asymptotic Coefficients for the Maximum Antichain of Partitions and Related Counting Inequalities
Dmitry Ignatov

The author considers open problems on counting (maximal) antichains in partition lattices and the size of maximum antichain. The new bounds for asymptotics coefficients obtained and the new integer sequence in OEIS encyclopedia.