Understanding Active Learning
Some of the fundamental concepts behind Active Learning.
I recently started an AI-focused educational newsletter, that already has over 150,000 subscribers. TheSequence is a no-BS (meaning no hype, no news etc) ML-oriented newsletter that takes 5 minutes to read. The goal is to keep you up to date with machine learning projects, research papers and concepts. Please give it a try by subscribing below:
The traditional concepts supervised learning imply that a model will be trained using a static and previously selected dataset of labeled data. That approach has resulted challenging as many of the most successful supervised learning techniques require very large amounts of data in order to achieve any learning milestones. Many times, building the best machine learning solutions comes down to the effectiveness of annotating unlabeled datasets. But how to prioritize? Give a set of unlabeled records, it is hard to predict which ones could have the biggest impact in the training of a given machine learning models. Welcome to the world of active learning?
