History

IDAO History

IDAO has been held since 2018. This page tells a story of the olympiad.

IDAO 2020

Competition

Stage 1. Online

 
This year the online task is coming from astronomy. It is focused on building a model that would predict the position of space objects using simulation data. The task was given by Russian Astronomical Science Center (ASC) and adopted for the Olympiad by the Laboratory of Methods for Big Data Analysis (LAMBDA, HSE University).Predicting the position of satellites is one of the most important tasks in astronomy. For example, information on the exact position of satellites in orbit is necessary to avoid extremely dangerous satellite collisions. Each collision leads not only to satellites destruction, but also results in thousands of space debris pieces. For instance, Iridium-Coscos collision in 2009 increased number of space debris by approximately 13%. Further collisions may result in Kessler syndrome and the inaccessibility of outer space. Also, a more accurate prediction of satellite position will help calculate more efficient maneuvers to save propellant and extend satellite life in orbit.

 


Leonid Gremyachikh
Task creator, Research Assistant at the Laboratory of Methods for Big Data Analysis (LAMBDA), HSE University

 

Online Stage results

Place

Track 1 Best Teams

Track 1 Score

Country

1

David&Sergey

96,96

Russia/ Switzerland 

2

Mrs. MIPT

96,57

France/ Switzerland 

3

Veni Vedi Vici

96,06

Russia/ Belarus

4

Snowflakes

95,95

Russia 

5

Extra Mile Stat

95,89

Peru

6

North people

95,29

Russia 

7-8

Selling gaRage

95,26

Russia

7-8

Baby Data O Plomo

95,26

France

9

oski

95,00

Russia

10

Gradient Ascent

94,82

Russia/ USA

11

The Land of Crimson Clouds

94,46

Russia

12

xenophon

94,20

Belgium

13

bestscraping

93,85

Russia

14

CHAD DATA SCIENTISTS

93,72

Russia

15

92,99

Russia 

 

Place

Track 2 Best Teams

Track 2 Score

Country

1

Data O Plomo

96,61

France

2

New Era los Guys

96,26

Russia 

3

QuMantumPhysicists

96,03

Japan

4

XSpace

95,87

Indonesia

5

IDA Pro

95,64

Russia

6

Earslaps Power

95,38

Russia

7

vrn

95,10

Russia

8

Maxibons

95,04

Spain

9

AKM

94,78

USA

10

Openprovider

94,22

Russia

11

random team

94,06

Switzerland

12

cutest_dog_in_the_ world

94,04

Russia

13

Hotteam v final

93,95

Russia

14

Alsetboost

93,95

Malaysia

15-16

EGS

93,93

Belarus/ Israel

15-16

Good Luck

93,93

Russia/ Kazakhstan

Stage 2. Finals

 
IMPORTANT!

You all have probably followed recent developments with COVID-19. Due to uncertainty caused by the pandemic, the Final will take place on November 21-25, 2020 and will be organized in the online format.

The winners of IDAO 2020 are:
🥇First place: Vrn (Ivan Bragin, Igor Kleynikov)
🥈Second place: Mylene Farmer (Ilya Ivanitskiy, Vasiliy Rubtsov, Anvar Kurmukov)
🥉Third place: random team (Ilya Kornakov, Kirill Borozdin)

Final Stage results
Place Team name Score Country
1 vrn 91.34 Russia
2 Mylene Farmer 91.18 Russia
3 random team 90.98 Russia/ Switzerland
4 TEAM X 90.96 Russia
5 Hotteam v final 90.95 Russia
6 CHAD DATA SCIENTISTS 90.82 Russia
7 Earslaps Power 90.78 Russia
8 QuMantumPhysicists 90.78 Japan
9 Baby Data O Plomo 90.75 France
10-11 Veni Vedi Vici 90.69 Belarus/ Russian
10-11 Openprovider 90.69 Kazakhstan/ Russian
12 Good Luck 90.63 Russia
13 The Land of Crimson Clouds 90.47 Russia
14 cutest_dog_in_the_world 90.44 Russia
15 David&Sergey 90.42 Russia/ Switzerland
16 Gradient Ascent 90.38 Russian/ USA
17 Zvezdochka** 90.36 Russia

 

Place Team name Score Country
18 xenophon 90.24 Belgium
19 IDA Pro 90.19 Russia
20 New Era los Guys 89.94 Russia
21 Selling gaRage 89.91 Russia
22 oski 89.75 Russia
23 Extra Mile Stat 89.74 Peru
24 Maxibons 87.46 Spain
25 XSpace 64.33 Indonesia
26 Baby Data O Plomo 61.18 USA
27 Alsetboost 57.82 Malaysia
28 North people 57.62 Russia
29-31 EGS 55.95 Belarus/ Israel
29-31 bestscraping 55.95 Russian
29-31 Mrs. MIPT 55.95 France/ Switzerland
32 Snowflakes 54.95 Russian
33 54.53 Russian

Platinum Partners

About QIWI plc. QIWI is a leading provider of next generation payment and financial services in Russia and the CIS. It has an integrated proprietary network that enables payment services across online, mobile and physical channels. It has deployed over 21.8 million virtual wallets, over 136,000 kiosks and terminals, and enabled merchants and customers to accept and transfer over RUB 116 billion cash and electronic payments monthly connecting over 44 million consumers using its network at least once a month. QIWI’s consumers can use cash, stored value and other electronic payment methods in order to pay for goods and services or transfer money across virtual or physical environments interchangeably.

 

Partner

 

Informational Partners

 

 

 

 

 

 

IDAO 2019: How it was?

previous arrow
next arrow
Slider
The final count is 1287 teams from 78 countries

Competition

Stage 1. On-line competition

There were two separate tracks during the online stage. From the machine learning perspective, the tracks were similar, yet the restrictions put on the solutions are different for each track.

The first track was a traditional data science competition. Having a labeled training data set, participants were asked to make a prediction for the test data and submit their predictions to the leaderboard. In this track, participants can produce arbitrarily complex models. If you like to use 4-level stacking or deep neural networks, this is the right track for you – you will only need to submit test predictions. However, those who qualify for the finals will be asked to submit the full code of the solution for validation by the judges.

In real world problems, efficiency is as important as quality. Complex and resource-intensive solutions will not fit the strict time and space restrictions often imposed by an application. That is why in the second competition track, the task was to solve the same problem as was in track one, but with tight restrictions on the time and on the memory used during both learning and inference. The participants had to upload the end-to-end code for your solution: both learning and inference. The evaluation server ran training and testing for the model and report the result. Both learning and evaluation must fit into time and memory constraints.

We hope that the two tracks made the olympiad fascinating for both machine learning competition experts and competitive programming masters, Kaggle winners and ACM champions, as well as everyone eager to solve real world problems with Data. Moreover, we encouraged people with different backgrounds, ML and ACM, to team up and push Data Analysis to new frontiers.

The muon research group of the LHCb experiment (LHCb Muon Group) provided the task for participants of the online qualifying round. Nikita Kazeev, co-author of the task, about the task:

“The task we gave the participants, muon identification, is important for the LHCb experiment. Majority of the physics research done at LHCb uses the output of this algorithm. I am looking forward to data science practitioners trying a hand on the problem. At the LHCb collaboration, we hope that the ideas and techniques they develop will ultimately bring us a step closer to understanding the big mysteries of the Universe. The task is also tricky from the machine learning point of view, for it contains features of variable length and negative example weights.”


Nikita Kazeev
PhD student at HSE and the University of Rome, researcher at the Laboratory of Methods for Big Data Analysis

Stage 2. Finals

The following two-step procedure has been used to select finalists.
Firstly, 15 teams with the highest score in the second track go to final (no matter what is their score in the first track).
Secondly, we consider all remaining teams and select 15 teams with the highest score in the first track (no matter what is their score in the second track).
These teams also went to the final.

Only submissions to the private tasks were considered.

Thus, in order to qualify for the final a team could choose one of the two strategies:

  1. to obtain the highest score in the second track where the code is needed, or
  2. to obtain the highest score in the first track.

Each of 30 teams, which were selected as finalists, received a letter describing further steps.
First of all, we ask the source code of your solution (for both tracks) which will be reviewed and validated. The solution must reproduce your submission exactly. Our jury members check that your solution contains no cheating, and your team does not attempt to unfairly pass the rules.
The finalists table 2019 was published in February after the jury’s decision.

The second, onsite stage was held in Moscow in April 2019 at the central headquarters of Yandex. Over the 36 hours of competition, participants tried not only to get up to speed on the model, but to create a full-fledged prototype that will be tested both in terms of accuracy and performance.

As part of the onsite round of the olympiad, speeches and workshops by international experts in machine learning and data analysis were also held.

Rules

To take part in the Olympiad, each team participant must register. Each team consists of 1-3 members.

The Olympiad is held in two rounds: online qualification round hosted on the Yandex.Contest Platform, and the on-site finals, held in Moscow. The solution of the task of the online round must be submitted by the team to the contest system no later than 23.59 Moscow time on February 11, 2019.

Based on the results of the online round, a table with points scored by teams will be published on the IDAO site by February 18, 2019, highlighting the list of finalists.

Each team can submit only one solution.

Only participants who have reached the age of 18 before the start of the on-site finals can participate.

At the finals, participants will need to use their own computer. Use of any legal software is allowed.

Three prizes will be awarded in the final round: one for the winning team, and two runners up.

Employees of Yandex and members of the LHCb collaboration can only participate hors concours, since Yandex and LHCb provide tasks for IDAO 2019.


Winners of the first stage (finalists) were invited to Moscow to take part in the on-site competition.

All participants have the chance to showcase their skills to the data science community on an international scale – the results will be internships, networking with some of the most passionate and like-minded individuals, and job opportunities. Winning also will be a serious advantage for students applying to the master’s degree programs at the HSE Faculty of Computer Science.

For winners, valuable prizes will be awarded. All members of the winning team will receive laptops as prizes. The winners will be determined by the leaderboard ranking based on private test set.

Finalists 2019

Team Captain First Member Second Member
AR_U_KIDDIN_MI Eugene Bobrov Moscow State University Vladimir Bugaevskii Moscow State University Denis Bibik Moscow State University
BarelyBears Hiroshi Yoshihara The University of Tokyo Kosaku Ono The University of Tokyo Naoki Maeda The University of Tokyo
Columbarium Konstantin Frolov SKB Kontur Grigoriy Pogorelov MTS Nikolay Prokoptsev Tinkoff Bank
DataBroom Pawan Kumar Singh Myntra Design Pvt. Ltd. Shruti Singh
DataScienceBois Egor Kravchenko Lomonosov Moscow State University Vladislav Trifonov Lomonosov Moscow State University Artyom Mironov Lomonosov Moscow State University
Eureka Sandeep Singh Adhikari Myntra Yadunath Gupta Myntra Nilpa Jha Myntra
FeelsBadMan Iskander Safiulin OKKO Ksenia Balabaeva ITMO Dmitry Ivanov Higher School of Economics
Gradient Boosting Pavel Shevchuk NRU HSE (applied mathematics) Mikhail Diskin NRU HSE Dmitriy Nikulin Samsung AI Center Moscow
HAL 9000 followers Aleksandr Belov National Research University of Electronic Technology, Applied mathematics Andrey Gorodetsky Bauman Moscow State Technical Maxim Tsygankov Bauman Moscow State Technical University
HardNet Yaroslav Murzaev MIPT Andrey Kachetov MIPT Viktor Nochevkin MIPT
holistic agency Maxim Shaposhnikov Ural Federal University Elena Arslanova Ural Federal University Denis Razbitsky Ural Federal University
Hunky-dory Ihar Shulhan Innopolis University Almira Murtazina Innopolis University Ruslan Mustafin Machine learning engineer
ifelse Samir Mammadov E-gov Development Center Asgar Mammadli E-gov Development Center Umid Suleymanov E-gov Development Center
ImprovY Stanislav Sopov SEMrush Mikhail Alekseev Okko Rinat Shakbasarov GrowFood
Inspiration Alexander Kolomoets UAC
itchy mcfly Petr Kuderov N/A Alex Maslov
John Keats Kirill Trofimov self-employed Sabina Abdullaeva
kek [1] Ranis Nigmatullin yandex
Livington Ivan Glebov MIPT
Magic CIty Sergei Arefev Saint Petersburg State University Artem Plotkin Saint Petersburg State University Roman Pyankov Saint Petersburg State University
Mylen Farmer Ilya Ivanitskiy Avito
Polis Yuriy Gavrilin Innopolis University Vladislav Kurenkov Innopolis University Andrey Kulagin Innopolis University
shadd Daniil Barysevich BSUIR, Computer Science Dzmitry Vabishchewich BSUIR Aliaksei Barysevich BSUIR, Computer Science
Singularis Lab Aleksei Alekseev Singularis Lab Oleg Shapovalov Singularis Lab Andrey Pedchenko Mello
TEAM X Andrey Kutsenko Moscow State University Nazar Beknazarov Higher School of Economics Sergey Kolomiyets Tyumen State University
Team_Name Daniil Cherniavskii MIPT Alexandr Valukov MIPT
trtr Denis Litvinov Sberbank Aleksey Buzovkin Michail Voronov
Umka Dmitrii Fedotov PJSC Norilsk Nickel
Unnamed:0 Arthur Bogdanov Innopolis University Gcinizwe Dlamini Innopolis University Rufina Galieva Innopolis University
wearenotgonnapasstothefinalanyways Toghrul Rahimli ADA University Jalal Rasulzade ADA University Orkhan Bayramli ADA University
Zvezdochka* Ernest Glukhov Innopolis University Daria Zapekina Innopolis University Vyacheslav Karpov Innopolis University

[1] The team “kek” participated in the final hors concours.

Winners

1st place – Mylen Farmer
Ilya Ivanitskiy – Higher School of Economics/Avito

2nd place – Zvezdochka*
Ernest Glukhov – Innopolis University,
Daria Zapekina – Innopolis University,
Vyacheslav Karpov – Innopolis University

3rd place – TEAM X
Andrey Kutsenko – Moscow State University,
Nazar Beknazarov – Higher School of Economics,
Sergey Kolomiyets – Tyumen State University

IDAO 2019 Finals Score in Yandex.Contest Ranks for Tasks Place by Rank
Team Name taskA taskB taskC Total Score Place by Score rankA rankB rankC Total Rank
Mylen Farmer 75.21 63.46 62.3 200.97 1 3.0 2.0 1.0 6.0 1
Zvezdochka* 75.38 62.91 61.8 200.09 2 2.0 6.0 5.0 13.0 2
TEAM X 74.91 62.55 62.09 199.55 4 5.0 8.0 2.0 15.0 3
shadd 74.66 62.85 62.06 199.57 3 9.0 7.0 3.0 19.0 4-5
BarelyBears 74.89 62.95 61.54 199.38 6 6.0 5.0 8.0 19.0 4-5
Eureka 73.93 63.45 62.04 199.42 5 15.0 3.0 4.0 22.0 6
Magic CIty 75.59 62.23 61.53 199.35 7 1.0 13.0 9.0 23.0 7
AR_U_KIDDIN_MI 73.96 63.27 61.77 199.0 8 14.0 4.0 6.5 24.5 8
Unnamed:0 73.8 63.65 61.46 198.91 9 18.0 1.0 11.0 30.0 9
FeelsBadMan 74.05 62.16 61.5 197.71 11 13.0 14.0 10.0 37.0 10
Team_Name 74.35 62.08 61.41 197.84 10 11.0 15.0 12.0 38.0 11
kek (hors concours) 74.1 62.39 61.04 197.53 12 12.0 10.0 16.0 38.0 12
HAL 9000 followers 74.7 62.32 60.26 197.28 13 7.0 11.5 27.0 45.5 13
Umka 73.83 61.93 61.28 197.04 15 17.0 16.0 13.0 46.0 14
itchy mcfly 71.95 61.88 61.77 195.6 20 23.0 18.0 6.5 47.5 15
ifelse 72.15 62.5 60.91 195.56 21 22.0 9.0 17.0 48.0 16
Gradient Boosting 74.98 61.6 60.55 197.13 14 4.0 25.0 20.0 49.0 17
Columbarium 74.64 61.87 60.52 197.03 16 10.0 19.0 22.5 51.5 18
Hunky-dory 70.76 62.32 61.16 194.24 24 25.0 11.5 15.0 51.5 19
Polis 74.67 61.78 60.45 196.9 17 8.0 20.0 25.0 53.0 20
Inspiration 73.87 61.89 60.52 196.28 18 16.0 17.0 22.5 55.5 21
Singularis Lab 72.24 61.73 61.24 195.21 23 21.0 22.5 14.0 57.5 22
ImprovY 73.54 61.63 60.52 195.69 19 19.0 24.0 22.5 65.5 23
wearenotgonnapass… 73.4 61.4 60.59 195.39 22 20.0 27.0 19.0 66.0 24
HardNet 68.31 61.75 60.85 190.91 27 27.0 21.0 18.0 66.0 25
holistic agency 71.13 61.57 60.52 193.22 25 24.0 26.0 22.5 72.5 26
trtr 68.98 61.73 60.35 191.06 26 26.0 22.5 26.0 74.5 27

Venue

The on-site finals, in which the top 30 performing teams from the online round will compete, has been held in Moscow, Yandex office.

Judges

  • Dmitry VetrovChairman of the Judiciary Commission, Research Professor in HSE, Head of the Deep Learning and Bayesian Methods Centre
  • Alexander GuschinJudge, Data Analyst at Yandex, highest overall rank in Kaggle is 5th
  • Emil KayumovJudge, Data Analyst at Yandex.Taxi
  • Matteo PalutanJudge, Researcher at the Laboratori Nazionali di Frascati of INFN, Member of the LHCb experiment at CERN
  • Barbara SciasciaJudge, Researcher at the Laboratori Nazionali di Frascati of INFN, Team leader of Frascati LHCb group and Deputy Operation Coordinator of the experiment
  • Evgeny SokolovJudge, Head of AI at Yandex.Zen, Deputy Head of the Big Data and Information Retrieval School
  • Dmitry UlyanovJudge, PhD student in Skoltech University, Research Scientist at Bayesian Methods Centre
  • Andrey UstyuzhaninJudge, Head of Methods for Big Data Analysis Lab at HSE

Dmitry Vetrov, Chairman of the judiciary commission, Research Professor in HSE, Head of the Deep Learning and Bayesian Methods Centre

Organizers

Yandex is a technology company that builds intelligent products and services powered by machine learning. Our goal is to help consumers and businesses better navigate the online and offline world. Since 1997, we have delivered world-class, locally relevant search and information services. Additionally, we have developed market-leading on-demand transportation services, navigation products, and other mobile applications for millions of consumers across the globe. Yandex, which has 22 offices worldwide, has been listed on the NASDAQ since 2011.

About our education initiatives: Yandex is helping shape the future of education by enhancing the learning process with machine learning technologies and teaching the next generation of data scientists to thrive in a world driven by artificial intelligence. As the leading search provider in Russia and one of Europe’s largest internet companies, we have a responsibility to help educate future generations in data science, artificial intelligence and machine learning. We are proud to help provide the education that will make this goal a reality and help future generations prepare for the jobs of tomorrow through our math and coding competitions, learning platforms, school programs, online courses, and the Yandex School of Data Analysis.

The Higher School of Economics (HSE) is the one of the most renowned Russian universities. The education is focused on economics and social sciences as well as high technologies and natural science. We stand on deep studying approach in fundamental disciplines combined with real experience at the biggest Russian companies to bring our graduates the perfect skills for their future carriers.

The HSE Faculty of Computer Science was created in March 2014 with the goal of becoming one of the world’s top 30 faculties in training developers and researchers in the field of big data storage and processing, system and software engineering and system programming. The Faculty is active in many research areas: machine learning, computer vision, theoretical computer science, algorithms for big data, optimisation, software engineering, and bioinformatics. We publish in leading computer science journals and present our results at major conferences.

Organizing Team

  • Tamara VoznesenskayaOrganizing Committee Chair, First Deputy Dean at the Faculty of Computer Science, HSE University
  • Irina PlisetskayaPartnership Coordinator, Deputy Dean for Development, Finance and Administration at the Faculty of Computer Science, HSE University
  • Sergey KarapetyanIDAO Coordinator, Manager at the Faculty of Computer Science, HSE University
  • Emil KayumovProblem Co-author, Data Analyst at Yandex.Taxi
  • Nikita KazeevProblem Co-author, PhD student at HSE and the University of Rome, researcher at the Laboratory of Methods for Big Data Analysis
  • Vladislav LipyaninWeb-Site Editor, Student at HSE University
  • Denis MashkovtsevSystem Administrator at HSE University
  • Alexey MitsyukTechnical Team Lead, Research Fellow at the Faculty of Computer Science, HSE University
  • Aleksey TolstikovYandex.Contest Expert, Yandex School for Data Analysis

Partners

Informational Partners

IDAO 2018: The first tournament

In 2018 the Online Round gathered 1500 participants from all over the world. It was conducted on January 15–February 11, 2018. 100 best participants took part in the On-Site Final that was held in Moscow on April 2–3.

previous arrow
next arrow
Slider

Competition

Stage 1. On-line competition

The event was organized by the HSE Faculty of Computer Science, Yandex, and Harbour.Space University (Barcelona) with the support of Sberbank. The task for the Online Round was provided by Yandex.Market.

Stanislav Fedotov, curator at the Yandex School of Data Analysis, and Associate Professor at the HSE Faculty of Computer Science, on the task for the Online Round: “At the online stage, the contestants solved a task for Yandex.Market. When a user enters this service with a specific purpose, the system chooses a set of options which match their query. For example, when someone looks for a kettle, Yandex.Market offers them a lot of options of kettles with various prices and options. But teaching the system to predict queries would be much more interesting, as this would mean that it would offer not what the individual is looking for at that particular moment, but something they would be likely to want in future. ‘The participants were given a search history of notional users, and they had to predict the categories of items these individuals hadn’t looked at over the last three weeks, but would be likely to search for in a week’s time. They had to choose five users, suggest five categories of goods for each user and ‘guess’ at least one of them

Stage 2. Final

The task for the Final was provided by Sberbank. According to Andrey Chertok, Managing Director for Research and Development at Sberbank, the participants had to solve a real problem on which the Sberbank team worked recently, and which is faced by all banks. The task is very applicable: it is about optimizing the cash supply for Sberbank ATMs, numbering tens of thousands across the country. The problem is that cash delivery isn’t always performed effectively, and as a result, cash lies useless in some ATMs, while others run out of cash too quickly. ‘The bank’s losses due to excessive money just ‘lying around’ in ATMs amounts to billions of roubles annually’, Andrey Chertok emphasized. ‘Our team uses data analysis more and more frequently to solve such problems. For example, the problem with cash delivery optimization and forecasting the amount of money to be cashed from a specific ATM was successfully solved with machine learning methods. We proposed a mini version of what we’ve done at Sberbank to the Olympiad participants.’ The finalists worked with real data of Sberbank ATMs’ locations and loading. During the process, the teams faced the same problems that are faced by bank data analysis teams in real life. This includes whether or not the data should be cleaned, and that the data sometimes has so-called ‘outliers’ which relate to more intensive cash delivery on days when salaries or pensions are paid. ‘In a short period of time, all the participants were quite successful in building usable models and got some hands-on experience in solving a real banking task’, said Andrey Chertok. ‘I believe, at this Olympiad, we managed to bring together competitive spirit and applicability’.

Winners 2018

In first place, and hailing from St. Petersburg, Magic City (Artem Plotkin, Roman Piankov and Sergey Arefev).

Coming in second place, all the way from Ukraine is team SantiagoSeaman (Alexander Makeev).

And finally in third place and making their way from Belarus, Apex (Evgeniy Demidovich, Sergei Petrov and Konstantin Mlynarchyk).

Partners

Informational and Tech Partners

Press about IDAO