May 13

Minsk 2017

What is #datafestby?

Data Fest is an informal conference that unites researchers, engineers and other data science folks.

  • No bullshit! Each speaker is a mature practitioner.
  • Broad coverage of topics from non-overlapping domains
  • Not just conference, but also a great party networking

Data Fest was held at Belarus Hi-Tech Park.

Photos

Schedule

  • 10:00 Introduction

    A short speech by DataFest partners and some organization announces.

  • Alex

  • 10:15 Make ML Great Again!

    Alex Natekin, Dictator at ODS.ai and Founder at DM Labs

    Warming up speeche by ODS.ai firekeeper

  • Eugene

  • 10:30 My path from Data Science newbie to Kaggle Master

    Eugene Babakhin, Data Scientist at Wargaming.net

    Overview of what Kaggle is, its pros and cons for the real life. Summary of competitions process and pipeline to handle them (with example for Kaggle competition: "Two Sigma Connect: Rental Listing Inquiries")

    Video | Slides

    Coffee break

  • Fedor

  • 11:30 How to build a self-driving car (in 100 lines of code)

    Fedor Chervinskii, Research Engineer at Yandex

    Overview of the technology stack inside a modern autonomous vehicle with focus on perception. Different approaches and problem formulations; deep learning for self-driving cars. Mediated perception: semantic segmentation, object detection, depth estimation. Training data: real and simulated

    Video | Slides

  • you!

  • 12:15 Free mic session

    The aim of this brief session is to expand your data science connections. It is an excellent opportunity for everybody to present themselves to the audience and say a couple of words on their projects and interests related to data science. The presentation format is a two-minute talk and a one-minute question-answer part.

    Lunch break

  • Artur

  • 14:00 Objects segmentation on satellite images (Kaggle DSTL Contest, 2nd place)

    Artur Kuzin, CV Analyst at Avito

    An overview of objects segmentation approaches on satellite images from Kaggle competition (Dstl Satellite Imagery Feature Detection). The speech will be dedicated to hacks and tricks of training and design deep convolutional neural networks collected from top-5 teams.

    Video | Slides

  • Alexey

  • 14:45 How machine learning helps to find effective prices for goods

    Alexey Chernobrovov, Director at Jet4Retail

    Review of approaches to pricing. How it was before and how machine learning will change it. The theory and practice. A few examples of the implementation of this approach at a large online stores.

    Video | Slides

    Coffee break

  • Anton

  • 15:45 ARIMA vs long term forecasts

    Anton Lebedevich, independent contractor

    Every textbook that has a time series chapter mentions ARIMA so it became a goto model for forecasting. But in real world it struggles with missing data, public holidays, non-stationarity, several steps ahead predictions. Anton will show better alternatives to ARIMA for long term forecasting.

    Video | Slides

  • Mike

  • 16:30 Explain me like I'm 5

    Mike Korobov, Software Engineer at ScrapingHub

    ELI5 is a Python library which allows to visualize and debug various Machine Learning models. It has built-in support for several ML frameworks and provides a way to explain black-box models. Mike is ELI5 core developer who will tell us more how this magic works.

    Video

    Coffee break

  • Petr

  • 17:30 From Jupyter notebook to production environment

    Petr Ermakov, Data Science and Big Data Team Lead at Mail.ru

    In his speech he will overview the long way from the proof of concept models to highload production maintenance.

    Video

  • Vitalii

  • 18:15 Sentiment analysis: best practices and challenges

    Vitalii Radchenko, Data Scientist at Ciklum

    Sentiment analysis is a very interesting task where there are many techniques which work well. We will cover data preprocessing, traditional ML, word- and char-based neural networks. Moreover, you will find out different tricks how to deal with small datasets, dataset absence and transfer learning.

    Video | Slides

    Beer party at Juno for active participants

    FAQ

    The FAQ section is incomplete but is being updated on a regular basis

    Do I have to register to attend Data Fest?

    You bet! If there are too many registrations, we will have to select N (best or random) candidates based on their questionnaires to be sure that participation will bring you maximum benefits.

    Are there any registration fees?

    DataFest is free for all participants. There are no registration charges. But you have to register.

    What if I got a refusal?

    Unfortunately, you will not be able to attend the event onsite. Still, don't be upset — we will have all our presentations broadcasted! Moreover, feel free to join in live discussion via the Open Data Science Slack channels

    When will I receive an event invitation?

    It will take some time for organizers to review all applications, but the process can be long, so please be patient. The first half of the invitations will be sent two weeks before the event and the rest approximately one week before the event start date.

    Will the conference materials be available later?

    Yes! The presentation videos and slides will be fully available in 3-4 weeks.

    So, you said all your speeches will be broadcasted?

    Yes, we plan to broadcast the speeches. A live webcast link will be available on our site.

    If there are not enough seats, what will be the application selection process?

    The main criteria we rely on are your experience in data science, as well as your motivation to attend the event.

    I still have questions!

    Contact a responsible guy via email me [at] arseny.info

    Partners