AI systems need to continually learn from new data to perform well in real-world scenarios. However, it is non-trivial to decide what new data needs to be labeled for training, and what is the best workflow and user interface for providing human feedback. This critical component of Machine Learning, called Active Learning, is often absent from Machine Learning courses. This Data Science Central webinar will extend TensorFlow's Deep Learning functionality with several Active Learning strategies, and apply these to the well-known ImageNet Computer Vision data set. At the end of this webinar you should be comfortable with combining your data annotation and Machine Learning strategies to continually improve your training data at scale.
Speaker: Robert Munro, VP of Machine Learning -- CrowdFlower
Hosted by: Bill Vorhies, Editorial Director -- Data Science Central