Active Learning by Acquiring Contrastive Examples

09/08/2021
by   Katerina Margatina, et al.
14

Common acquisition functions for active learning use either uncertainty or diversity sampling, aiming to select difficult and diverse data points from the pool of unlabeled data, respectively. In this work, leveraging the best of both worlds, we propose an acquisition function that opts for selecting contrastive examples, i.e. data points that are similar in the model feature space and yet the model outputs maximally different predictive likelihoods. We compare our approach, CAL (Contrastive Active Learning), with a diverse set of acquisition functions in four natural language understanding tasks and seven datasets. Our experiments show that CAL performs consistently better or equal than the best performing baseline across all tasks, on both in-domain and out-of-domain data. We also conduct an extensive ablation study of our method and we further analyze all actively acquired datasets showing that CAL achieves a better trade-off between uncertainty and diversity compared to other strategies.

READ FULL TEXT
research
11/27/2020

Deep Active Learning for Sequence Labeling Based on Diversity and Uncertainty in Gradient

Recently, several studies have investigated active learning (AL) for nat...
research
07/31/2023

Active Learning in Genetic Programming: Guiding Efficient Data Collection for Symbolic Regression

This paper examines various methods of computing uncertainty and diversi...
research
06/20/2022

Actively Learning Deep Neural Networks with Uncertainty Sampling Based on Sum-Product Networks

Active learning is popular approach for reducing the amount of data in t...
research
09/22/2020

ALICE: Active Learning with Contrastive Natural Language Explanations

Training a supervised neural network classifier typically requires many ...
research
06/25/2022

Making Look-Ahead Active Learning Strategies Feasible with Neural Tangent Kernels

We propose a new method for approximating active learning acquisition st...
research
12/04/2020

Bayesian Active Learning for Wearable Stress and Affect Detection

In the recent past, psychological stress has been increasingly observed ...
research
04/14/2020

Contrastive Examples for Addressing the Tyranny of the Majority

Computer vision algorithms, e.g. for face recognition, favour groups of ...

Please sign up or login with your details

Forgot password? Click here to reset