Active Learning Survey

Source: Deep Learning on Medium

Active Learning Survey

One of the last active learning survey articles was published in 2009 (Burr Settles; updated in 2010). The purpose of this note is to highlight some of the recent advances in the field of active learning and provide updates Settles’s article. This not is not meant to be exhaustive and is a work in progress.

For each method, we provide i) a short summary of the method and its novelty, ii) comments on the code (if available), and iii) potential next steps.

“Evaluating Uncertainty Quantification in End-to-End Autonomous Driving Control”

“FAST AND SCALABLE BAYESIAN DEEP LEARNING BY WEIGHT-PERTURBATION IN ADAM”

“ACTIVE LEARNING FOR CONVOLUTIONAL NEURAL NETWORKS: A CORE-SET APPROACH”