The Microsoft Machine Learning and Intelligence School will take place in Saint Petersburg, Russia, from July 29 to August 5, 2015. The School is sponsored by Microsoft Research and Yandex and is organized in cooperation with Lomonosov Moscow State University (MSU). The school offers advanced undergraduates, PhD students, and young scientists and developers a unique opportunity to learn about the latest research in machine learning, intelligence and data science from top scientists. The Machine Learning and Intelligence School 2015 will bring together thought leaders and practitioners from a broad range of disciplines including computer science, engineering, statistics, mathematics and economics. Participants will learn some of the key challenges posed by this new era of machine learning and intelligence, and will work on the next generation of approaches, techniques and tools that will be needed to exploit the information revolution for the benefit of society. In this context, we are organizing a bootcamp to enter the AutoML challenge.
The challenge is to beat the results of ROUND 1 of the AutoML challenge. To that end, we set up a special website clone.
First day (Tweakathon 1): Venus, Mercury, Ceres
Last day (Final 1): Venus, Rhea, Ceres
Successful submission in AutoML 1.5: Saturn
Best presentation: Mercury
Thurday, July 30, 2015
16:15: Introduction, Isabelle Guyon. [SLIDES]
16:45 - 17:30: Python tutorial (1). Olga Liakhovich. [iPython notebook]
Friday, July 31
10:00 - 11:00: AutoML tutorial. Isabelle Guyon. [SLIDES]
11:15 - 12:30: Python tutorial (2). Olga Liakhovich. [iPython notebook]
12:30 - 13:00: Hackathon instructions. Lukasz Romaszko, Arthur Pesah.
14:00 - 17:30: Hackathon on round 1 data.
17:30 - 18:00: Hackathon provisional winners on validation data.
Monday, August 3
16:15 - 18:00: Hackathon on round 1 data (continued).
Tuesday, August 4
17:15 - 18:00 : Final submissions, including 1 slide presentation.
Wednesday, August 5
15:30 - 16:30 : Hackathon student presentations. [SLIDES]
The participants are organized in groups. The group name is the Codalab ID. It is possible to communicate with the members of the group by writing to <group-name>@chalearn.org.