Diversity Enhanced Active Learning with Strictly Proper Scoring Rules
Dec 6, 2021
Speakers
About
We study acquisition functions for active learning (AL) for text classification. The Expected Loss Reduction (ELR) method focuses on a Bayesian estimate of the reduction in classification error, recently updated with Mean Objective Cost of Uncertainty (MOCU). We convert the ELR framework to estimate the increase in (strictly proper) scores like log probability or negative mean square error, which we call Bayesian Estimate of Mean Proper Scores (BEMPS). We also prove convergence results borrowing techniques used with MOCU. In order to allow better experimentation with the new acquisition functions, we develop a complementary batch AL algorithm, which encourages diversity in the vector of expected changes in scores for unlabelled data. To allow high performance text classifiers, we combine ensembling and dynamic validation set construction on pretrained language models. Extensive experimental evaluation then explores how these different acquisition functions perform. The results show that the use of mean square error and log probability with BEMPS yields robust acquisition functions, which consistently outperform the others tested.We study acquisition functions for active learning (AL) for text classification. The Expected Loss Reduction (ELR) method focuses on a Bayesian estimate of the reduction in classification error, recently updated with Mean Objective Cost of Uncertainty (MOCU). We convert the ELR framework to estimate the increase in (strictly proper) scores like log probability or negative mean square error, which we call Bayesian Estimate of Mean Proper Scores (BEMPS). We also prove convergence results borrowing…
Organizer
NeurIPS 2021
Account · 1.9k followers
About NeurIPS 2021
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
OSOA: One-Shot Online Adaptation of Deep Generative Models for Lossless Compression
Watch later
Chen Zhang, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Learning to Assimilate in Chaotic Dynamical Systems
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Combining Latent Space and Structured Kernels for Bayesian Optimization over Combinatorial Spaces
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Near-Optimal Multi-Perturbation Experimental Design for Causal Structure Learning
Watch later
Scott Sussex, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Robust Inverse Reinforcement Learning under Transition Dynamics Mismatch
Watch later
Luca Viano, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Efficient and Local Parallel Random Walks
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%