IT City, the largest regional conference for entrepreneurs, managers and professionals working in information technology, was held September 6-7 in Tomsk, Russia.
This year NTR hosted a two hour machine learning session. We shared our experience developing recommendation systems, talked about trends and the development of artificial intelligence systems over the past year, and about experiments with an artificial neural network.
There were four speakers plus a panel discussion at the end.
Nick Mikhailovsky, NTR’s CEO
Headliner
“How a good recommendation system works”.
Theses:
DASE Architecture – “MVC Machine Learning”
Offline and online assessment of the quality of the recommendation system
Anatomy of a personalized recommendation
Bio
CEO and co-founder of NTR Group. Investor, board of directors of Sinergo, a successful system integrator working in the mining industry, co-founder of several startups, including Crossss, a recommendation service for online stores.
Graduate of the Moscow Institute of Physics and Technology with a degree in Applied Mathematics and graduate school in Matrix Analysis. Technical background in the areas of: artificial intelligence / predictive analytics, end-to-end processing, information system architecture, and information retrieval systems. Worked on neural networks before they became mainstream.
We invited two additional speakers who work in Moscow-Silicon Valley startups and one from Tomsk – the Netherlands company.
Grigory Sapunov, Сo-Founder & CTO, Intento
“AI – Last Year’s Progress (2018-2019)”
Theses included:
1. Photorealistic content generation. 2. Natural Language Processing (NLP). 3. Speech & Sound. 4. AI Strategies. 5. Turing Award, the Nobel Prize of Computing. 6. Artificial General Intelligence. 7. Games. 8. Medicine & Biology. 9. Large-scale real-life applications. 10. Cloud AI. 11. Edge AI.
Bio
ML/DL/AI expert. Software engineer with 20+ years programming experience. Loves Life Sciences. CTO and co-founder of Intento. Google Developer Expert in ML.
Mikhail Salnikov
Data Scientist, Wallarm
Theses:
Traditional IPS / IDS solutions, including WAF, produce so many false positives that they are almost impossible to manage. The reason for this lies in the fact that they are based on outdated detection mechanisms, such as signatures, regular expressions, etc.
In this presentation I will talk about the neural network approach for detecting attacks on which the Wallarm platform is based, including a detailed description of the implementation of the model for detecting false positives.
Bio
Analyst-Algorithmist. A professional Data Scientist specializing in sequence processing tasks. Mikhail works in Valarm and solves the problems of automatically detecting web attacks simultaneously studying for his Master’s in Data Science at Skoltech.
Dmitry Krasavin
Data Scientist, ENBISYS
“Gray box” or how to get more than just the number at the output from the ANN
Theses:
How to understand what interests the nets?
We analyze the example of medical images, time series and ImageNet samples.
How to get confidence intervals in regression tasks?
Consider the example of obtaining a confidence interval for assessing the value of real estate.
How to teach the nets to doubt dubious input?
Let us compare the reactions of ordinary and Bayesian neural networks to input data from a distribution other than the training set.
Bio
Machine Learning course lecturer at Tomsk State University
IT City 2018 Speaker – “Another step towards the future or how a neural network teaches Dutch children”
Image source: IT City
Left to right: Tatiana (HR), Leonid (Engineer), Daria (me, marketing), Aleksandr (PM).
Two more members of NTR’s staff were invited to participate at the conference.
Tatiana Nikolakhina, NTR HR Director, was invited to judge the themed game “Management Struggle.” The game is used for negotiation practice. Four participants conduct negotiations on six specific conflict situations that are regularly encountered in our work. Resolution is necessary to achieve the goals specified in the task. The winner was decided by five judges chosen from among the heads of Tomsk companies.
Max Ivankin, NTR Project Manager, was invited as an expert to the Project Management session. There was a battle royal on hot topics; so heated that the experts decided to meet again soon to delve further into the subjects.
Our session was very successful. It was well attended and we liked being speakers and hosting this kind of trending session. It was a good experience for our company. See you at the next IT conference.
Last year’s post: Tomsk – IT city in Russia.
Tell a friend!