Streaming Active Learning with Deep Neural Networks

2023/03/05 by Akanksha Saran, Saran, Akanksha, Safoora Yousefi +7 · 5 citations
Computer Science · #Machine Learning and Algorithms #Data Stream Mining Techniques #Machine Learning and Data Classification

paper · pdf · doi:10.48550/arxiv.2303.02535

Abstract

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.

Cited by

• brat: Aligned Multi-View Embeddings for Brain MRI Analysis
• Towards Active Synthetic Data Generation for Finetuning Language Models
• Representation-Based Exploration for Language Models: From Test-Time to Post-Training
• Towards Multimodal Active Learning: Efficient Learning with Limited Paired Data
• TAPS : Frustratingly Simple Test Time Active Learning for VLMs

Related

• Deep Active Learning: Unified and Principled Method for Query and Training
• Deep Active Learning over the Long Tail
• PAL : Pretext-based Active Learning
• Improved Algorithms for Neural Active Learning
• Accelerating Batch Active Learning Using Continual Learning Techniques