Streaming Active Learning with Deep Neural Networks
This work addresses the challenge of realistic active learning scenarios, such as in HCI and large datasets, by enabling streaming applications, though it is incremental as it builds on prior batch active learning methods.
The paper tackles the problem of applying deep neural networks to active learning in streaming settings, where data arrives sequentially, and proposes VeSSAL, an algorithm that queries groups of points for labels upon encounter, achieving a desired query rate without hand-tuned hyperparameters.
Active learning is perhaps most naturally posed as an online learning problem. However, prior active learning approaches with deep neural networks assume offline access to the entire dataset ahead of time. This paper proposes VeSSAL, a new algorithm for batch active learning with deep neural networks in streaming settings, which samples groups of points to query for labels at the moment they are encountered. Our approach trades off between uncertainty and diversity of queried samples to match a desired query rate without requiring any hand-tuned hyperparameters. Altogether, we expand the applicability of deep neural networks to realistic active learning scenarios, such as applications relevant to HCI and large, fractured datasets.