Follow our latest advances in machine learning.

Computer vision

Distributional shift

Generative models

Graph machine learning

Machine learning theory

Nearest neighbor search

Uncertainty estimation 

Bayesian methods

Gradient boosting

Natural language processing 

Tabular data

Blog

Yandex Research Blog

Graphs are a natural way to represent data from various domains such as social networks, molecules, text, code, etc. We develop and analyze algorithms for graph-structured data.

We have proposed a benchmark of five diverse heterophilous graphs that come from different domains and exhibit a variety of structural properties. Our benchmark includes a word dependency graph roman-empire, a product co-purchasing network amazon-ratings, a synthetic graph emulating the minesweeper game minesweeper, a crowdsourcing platform worker network tolokers, and a question-answering website interaction network questions.

Introducing new heterophilous graph datasets

Four papers by the Yandex Research team have been accepted for publication at the International Conference on Learning Representations (ICLR 2023).

Papers accepted to ICLR 2023

Tabular data involves two-dimensional tables with objects (rows) and features (columns), which are used in numerous applied tasks such as classification, regression, ranking and many others.

In this post, we share exciting findings from our recent NeurIPS 2022 paper on deep learning and tabular data. We discuss how improved representations of numerical features can turn your vanilla MLPs (multilayer perceptrons) and Transformers into state-of-the-art tabular deep learning solutions that are competitive even with gradient-boosted decision trees, one of the strongest baselines in the world of tabular data problems!

Embeddings for numerical features in tabular deep learning

Two papers have been accepted for publication at the Conference on Empirical Methods in Natural Language Processing (EMNLP 2022).

Papers accepted to EMNLP 2022

large_Graph-based NNS.png

medium_Graph-based NNS.png

small_Graph-based NNS.png

thumbnail_Graph-based NNS.png

Graph-based NNS.png

Nearest neighbor search is a long-standing problem arising in a large number of machine learning applications, such as recommender services, information retrieval, and others.

Nearest Neighbor Search (NNS) is a long-standing problem arising in many machine learning applications, such as recommender services, information retrieval and others. The goal of the nearest neighbor search is to find the closest element to a given query vector within a database of n elements. Since the sizes of databases in practical tasks are constantly increasing, the scalability of NNS becomes crucial.

Efficient NNS receives much attention from the machine learning community. The problem is well-solved for low-dimensional spaces. However, most practical applications need effective and accurate search in high dimensional ones. Due to the curse of dimensionality, for large d the algorithms for exact NNS provide little improvement over a brute-force algorithm. Thus, both researchers and practitioners are interested in approximate solutions which are efficient and quite accurate.

Recently, methods based on similarity graphs were shown to outperform tree-based and LSH-based techniques according to the public benchmarks. In this post, we discuss graph-based methods from different perspectives. In particular, we cover their scalability, applicability to a wide range of applications, and theoretical properties.

Blog

Posts

Introducing new heterophilous graph datasets

Papers accepted to ICLR 2023

Embeddings for numerical features in tabular deep learning

Papers accepted to EMNLP 2022

Graph-based nearest neighbor search

Filter by: