Open in app

Sign In

Write

Sign In

VK Team
VK Team

480 Followers

Home

About

Jan 31

Avoiding loss of money: notifications of drops in product metrics

Contents: Finding the best approach Current vs. last period Moving average Moving boundaries ML methods Our algorithm Interface Redash Email notifications Telegram bot Hourly notifications Further improvements Conclusions Trying to keep track of all different kinds of metrics and snapshots on dashboards, you can easily miss an important change in metrics…

Analytics

11 min read

Avoiding loss of money: notifications of drops in product metrics
Avoiding loss of money: notifications of drops in product metrics
Analytics

11 min read


Jan 9

Contextual multi-armed bandits for content recommendation, or not by Bernoulli alone

Why the “classic” ML recommendations are lacking, how to promote content on social media using multi-armed bandits, and how to take into account the context in this process. My name is Alexander Sukhochev, I work in machine learning and lead the team handling recommendations and development of VK services. I…

Machine Learning

22 min read

Contextual multi-armed bandits for content recommendation, or not by Bernoulli alone
Contextual multi-armed bandits for content recommendation, or not by Bernoulli alone
Machine Learning

22 min read


Jun 28, 2022

VK’s Face in 2022: New Icons, Logos, Colours, and Style

We had a global rebrand last October: Mail.ru Group turned into VK. This article tells about our ideas and changes. Many are aware of the brands within VK ecosystem: Mail.ru, social networks VK and OK, VK Play, GeekBrains, Skillbox, and others. …

Design

5 min read

VK’s Face in 2022: New Icons, Logos, Colours, and Style
VK’s Face in 2022: New Icons, Logos, Colours, and Style
Design

5 min read


Feb 18, 2022

Pairing Git with Webpack+React application

Sometimes showing some information from your git repository right in your application can be helpful. My name is Nikolay Syrmolotov, I’m a front-end engineer at VK Pay and in this article, we’re going to take advantage of the NodeJS built-in function execSync and display three versions of the main branch…

Webpack

4 min read

Pairing Git with Webpack+React application
Pairing Git with Webpack+React application
Webpack

4 min read


Feb 10, 2022

Read Instead of Listen: How Speech Recognition Works on VKontakte

When it comes to messages, reading is faster than listening. It’s also easier to skim through text to find and verify details. However, sometimes there are situations where it’s a lot more convenient to just send a voice message than type it all out. My name is Nadya Zueva. In…

Asr

12 min read

Read Instead of Listen: How Speech Recognition Works on VKontakte
Read Instead of Listen: How Speech Recognition Works on VKontakte
Asr

12 min read


Oct 11, 2021

Neural Networks vs. Vacuum Cleaners, or How We Denoised VK Calls

These days, what other way is there to implement noise suppression other than neural networks? But we aren’t here to just point out the obvious and disappear into thin air. My name is Vitaly Shutov, I’m a Machine Learning Engineer at VK.com and in this article we’ll talk about the…

Ds

12 min read

Neural Networks vs. Vacuum Cleaners, or How We Denoised VK Calls
Neural Networks vs. Vacuum Cleaners, or How We Denoised VK Calls
Ds

12 min read


Aug 12, 2021

A Little Journey of Raw Events: From Logs to Experimental Results

Introduction In recent years, there have been many papers on A/B experiments published. The vast majority of them are about statistical tests and their usage limitations. However, in this article, we are going to look at the entire process of such experiments, including data pipelines, specific data structures, and statistical procedures. …

24 min read

A Little Journey of Raw Events: From Logs to Experimental Results
A Little Journey of Raw Events: From Logs to Experimental Results

24 min read


Jul 12, 2021

PYMK at VK: ML over EGO-NETS

The ability to add users as friends is one of the most important mechanics of any social network. The vast majority of interactions occur between users who are friends with each other. They see and comment on each other’s posts in their news feeds and go to their friends lists…

Machine Learning

10 min read

PYMK at VK: ML over EGO-NETS
PYMK at VK: ML over EGO-NETS
Machine Learning

10 min read


May 27, 2021

Data Debugging in Collaborative Filtering with Implicit Feedback: Does It Work?

Hi, my name is Daniel. I’m a member of the Core ML Team at VK.com. I work on recommender systems, which is also the topic of this post. An interesting paper was presented at NeurIPS 2020: https://papers.nips.cc/paper/2020/hash/019fa4fdf1c04cf73ba25aa2223769cd-Abstract.html. The main idea of it is that if you somehow preprocess the training…

8 min read

Data Debugging in Collaborative Filtering with Implicit Feedback: Does It Work?
Data Debugging in Collaborative Filtering with Implicit Feedback: Does It Work?

8 min read


Sep 11, 2020

Causal inference pipeline for VK metrics

Hi, everyone. I’m Anver from the CoreML team at VK, where I design and improve algorithms for news feed ranking. Improving ranking algorithms usually results in improving the click-through rates(CTRs) of users’ actions (e.g., CTR of likes, CTR of comments, etc.). We will refer to these metrics as short-term metrics. …

Causal Inference

10 min read

Causal inference pipeline for VK metrics
Causal inference pipeline for VK metrics
Causal Inference

10 min read

VK Team

VK Team

480 Followers

About VK technologies and infrastructure

Help

Status

Writers

Blog

Careers

Privacy

Terms

About

Text to speech