Google I/O Extended events help developers from around the world take part in the I/O experience. This year, during Google I/O event, UC Santa Barbara Computer Science Professor William Wang gave a keynote speech at the Santa Barbara extended event (hosted by LogMeIn) on the topics of reinforcement learning and semi-supervised learning.

In particular, Prof. Wang introduced the reinforced co-training algorithm that he designed together with Ph.D. student Jiawei Wu. Co-training is a popular semi-supervised learning framework to utilize a large amount of unlabeled data in addition to a small labeled set. Co-training methods exploit predicted labels on the unlabeled data and select samples based on prediction confidence to augment the training. However, the selection of samples in existing co-training methods is based on a predetermined policy, which ignores the sampling bias between the unlabeled and the labeled subsets, and fails to explore the data space. The team proposes a novel method, Reinforced Co-Training, to select high-quality unlabeled samples to better co-train on. More specifically, this approach uses Q-learning to learn a data selection policy with a small labeled dataset, and then exploits this policy to train the co-training classifiers automatically. Experimental results on clickbait detection and generic text classification tasks demonstrate that this proposed method can obtain more accurate text classification results.