Improved Algorithms for Neural Active Learning
This work addresses the challenge of efficient active learning for k-class classification problems, offering incremental improvements in theoretical and empirical performance over existing methods.
The paper tackles the problem of improving neural-network-based active learning algorithms in non-parametric streaming settings by introducing new regret metrics and a tailored algorithm, resulting in an instance-dependent regret upper bound that improves by a multiplicative factor O(log T) and removes the curse of input dimensionality, with empirical validation showing enhanced performance.
We improve the theoretical and empirical performance of neural-network(NN)-based active learning algorithms for the non-parametric streaming setting. In particular, we introduce two regret metrics by minimizing the population loss that are more suitable in active learning than the one used in state-of-the-art (SOTA) related work. Then, the proposed algorithm leverages the powerful representation of NNs for both exploitation and exploration, has the query decision-maker tailored for $k$-class classification problems with the performance guarantee, utilizes the full feedback, and updates parameters in a more practical and efficient manner. These careful designs lead to an instance-dependent regret upper bound, roughly improving by a multiplicative factor $O(\log T)$ and removing the curse of input dimensionality. Furthermore, we show that the algorithm can achieve the same performance as the Bayes-optimal classifier in the long run under the hard-margin setting in classification problems. In the end, we use extensive experiments to evaluate the proposed algorithm and SOTA baselines, to show the improved empirical performance.