All videos
Opera view

The reasons we do not do Machine Learning any more.

October 8, 2019

Last year I was talking how back at AirHelp we use machine leaning. As an example I have used the classical problem: analyzing the pictures with convolutional neural networks. This season I'd like to tell you when we decide that the ML approach is the wrong choice.I will share the story of what has happened to this CNN we have created last year.

Tags

Data Machine Learning

About the speaker

Michał Jakóbczyk

Michał Jakóbczyk

Software Engineering Director at Airhelp
Primarily I'm a Software Engineer. For last fifteen years doing Big Data, data analysis, machine learning, product and project management. For last several years focused…
Read more

Other videos that you might like

fallback image
Data
Machine Learning in Visual Search
Aleksandra Kunysz
fallback image
Data
Is your phone ready to do machine learning?
Adam Niedziałkowski
The reasons we do not do Machine Learning any more.
Data
pandas-stubs — How we enhanced pandas with type annotations
Joanna Sendorek, Zbyszek Królikowski
fallback image
Data
From Spark MLlib model to learning system with Watson Machine Learning
Łukasz Ćmielowski
All videos