Machine Learning Boot Camp IV. Fourth. Secret. Your

    image


    On April 21, we are opening the fourth machine learning competition on the ML Boot Camp platform. Today we will talk about a new task, updates on the site and other useful nishtyak. And if you suddenly hear for the first time what ML Boot Camp is, go under the spoiler and we will tell you everything.


    About ML Boot Camp

    ML Boot Camp is a platform for solving machine learning tasks. We periodically post new tasks on it and launch a contest. Participants must solve our problem within a month and send a solution. Authors of the best solutions will receive prizes. In the last championship, we gave the MacBook Air first place, the iPad second and third, and iPod nano 4-6 places.


    At the start, participants receive the conditions of the task, a verbal description of the available data - a training sample. The sample consists of labeled examples - description vectors of each object with a known answer. Participants use computer-known methods of machine learning to train a computer. They use the trained system on new objects (test sample), trying to determine the answer for them.


    The test sample is randomly divided into two parts: rating and final. The overall result on the rating data is calculated by the system and published immediately, but the winner is the one who gets the best results on the final data. The results remain hidden to the participants until the very end of the competition.


    On the last day of the championship, the participant can choose two decisions that will represent him in the final. The best of them will count towards the leaderboard.


    New challenge


    This time we offer you the "Task with a secret." We will not disclose a meaningful statement of the problem. She will remain unknown until the end of the competition. You will be able to test your analytical skills in full!


    You are faced with the task of classification: based on the well-known distribution of the five classes of teaching elements, distribute the test ones. As a response, send a text file, each line of which corresponds to a line in the file with test data and contains the class number (0, 1, 2, 3 or 4). We offer you as many as 42 numerical signs for classification!


    The criterion for the quality of the solution will be the proportion of correctly classified objects. The test sample is randomly divided into two parts in a ratio of 40/60. The result at the first 40% will determine the position of participants in the rating table throughout the competition. The result for the remaining 60% will be known after the end of the competition and determine the final arrangement of the participants. Good luck


    We express our gratitude to the UNN to them. N. I. Lobachevsky and personally Nikolai Zolotikh and Oleg Durandin for their help in preparing the task and expert support for the championship! Nikolai and Oleg participated in holding each of our ML championships, without them we would not have mastered half of what has been done now.


    Useful materials


    Educational article


    If you are a beginner, we recommend that you read the small tutorial on our platform. In it, you will analyze the "Credit Scoring" task and learn how to predict whether the loan will return to the bank according to the client.


    image
    The article contains squeezed test data, its visualization, pieces of Python code and all semantic conclusions


    Parsing ML BootCamp I


    At ML Boot Camp, we already worked with anonymous data. In a closed student contest, we asked to classify binary sequences. The proportion of correct answers was also a quality criterion. Pavel Shvechikov achieved an impressive 0.6785, provided that some sequences were written by people, the second - a random number generator, and the third - an algorithm.


    We asked the winners of the contest to tell the main ideas of their solution and collected them in a separate publication on Habré . Perhaps their ideas will help you choose the direction of movement. Look, there are cool visualizations there:


    image
    Everything can be visualized. Even binary sequences


    Sandbox


    You can practice before the start of the championship, including on the “Binary Trees” task in the Sandbox . Any tasks of past championships are available there, you can download your solution and find out the score. For each task, the sandbox has its own leaderboard. If the new task seems to you too difficult (or, conversely, simple) - conquer the rest.


    image
    In the sandbox you can solve all the tasks of the old contests


    Chat in Telegram


    Now, thanks to the official championship chat you can ask your question directly to the organizers. You can also ask for advice or share guesses about the decision. All participants gather here and storm the task. You will be helped with fresh ideas and kind words.


    image
    Experienced machine operators participate in the chat, including winners of past contests.


    Forum


    The participants very much requested a forum for putting out something that could easily be lost in the chat. It’s not a fact that we will manage to open it just before the start of the competition, but we can certainly promise that in the near future the forum will appear on our site.


    Two solutions as an answer


    From now on, you can choose two solutions as your final answer. The one that gets the highest Score in the final sample will be your result in the championship. This will help you, for example, if in one of the solutions you have a more stable model, but the other gives the best result in the test sample.


    Prizes


    This time we will break the slender ranks of Apple equipment in the prize pool. For first place we’ll give a MacBook Air 13 laptop , for second and third - a smart watch Samsung Gear S3 Frontier . If you do not get into the top three, but enter the TOP-6, your personal WD My Cloud 6TB cloud drive will make friends with you . And, as always, the TOP-50 participants of the championship will receive T-shirts with the championship logo.


    registration


    The championship will open on April 21 at 14:00 Moscow time. You can register on the platform at this link . Until the movement begins, come to solve problems in the Sandbox .


    Also popular now: