BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learning
BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learning
复制标题
DOI：
--
复制
发表时间：
2019-06
期刊：
影响因子：
--
通讯作者：
Andreas Kirsch;Joost R. van Amersfoort;Y. Gal
中科院分区：
文献类型：
--
作者：
Andreas Kirsch;Joost R. van Amersfoort;Y. Gal
我们开发了BatchBALD，一种对一批点和模型参数之间的互信息的易于处理的近似，我们将其用作获取函数，以联合选择多个信息点来执行深度贝叶斯主动学习任务。BatchBALD是一种贪婪的线性时间$1-\FRAC{1}{e}$-近似算法，适用于动态规划和高效缓存。我们将BatchBALD与常用的批量数据采集方法进行了比较，发现当前的方法获取了相似和冗余的点，有时性能不如随机采集数据。最后，我们展示了，使用BatchBALD考虑采集批次中的相关性，我们在标准基准测试中实现了最新的性能，大幅提高了批量采集的数据效率。
We develop BatchBALD, a tractable approximation to the mutual information between a batch of points and model parameters, which we use as an acquisition function to select multiple informative points jointly for the task of deep Bayesian active learning. BatchBALD is a greedy linear-time $1 - \frac{1}{e}$-approximate algorithm amenable to dynamic programming and efficient caching. We compare BatchBALD to the commonly used approach for batch data acquisition and find that the current approach acquires similar and redundant points, sometimes performing worse than randomly acquiring data. We finish by showing that, using BatchBALD to consider dependencies within an acquisition batch, we achieve new state of the art performance on standard benchmarks, providing substantial data efficiency improvements in batch acquisition.