Vidora CTO Abhik Majumdar Describes the Machine Learning Pipeline

Last month, Vidora held ‘The Future of Machine Learning Business’ event in conjunction with News Corp, where we discussed where the machine learning space is headed, where the latest innovations are most likely to come from, and how media businesses such as News Corp are thinking about ML and planning for its future.

Part One of the event was a talk by Vidora CTO Abhik Majumdar, PhD on the future of machine Learning in business. In this clip,  Abhik describes the various steps required to implement a ML algorithm and shows that deploying ML algorithms within businesses is much more difficult than just choosing the best neural network model. In particular, he illustrates through examples that the hardest parts of machine learning actually lie in the initial steps of the pipeline, such as feature cleaning and feature engineering. It is in these areas where Vidora believes the biggest innovations within ML will take place over the coming years.

Click here to see a segment from Abhik’s speech about the Future of ML, and reach out to us at to learn more!

Posted by Ajai Vishwanath on October 29, 2018