Adaptive regularization of weight vectors
Adaptive regularization of weight vectors
复制标题
DOI：
10.1007/s10994-013-5327-x
复制
发表时间：
2013-05-01
期刊：
影响因子：
7.5
通讯作者：
Dredze, Mark
中科院分区：
文献类型：
--
作者：
Crammer, Koby;Kulesza, Alex;Dredze, Mark
我们提出了AROW，一个在线学习算法的二进制和多类问题，结合了大幅度的培训，置信加权，并能够处理不可分离的数据。AROW在看到每个新实例时执行预测函数的自适应正则化，使其在存在标签噪声的情况下表现得特别好。我们推导出错误的二进制和多类设置的形式类似的二阶感知器界。我们的边界不假设可分性。我们还将我们的算法与最近的置信加权在线学习技术。实证评估表明，AROW实现了最先进的性能在广泛的二进制和多类任务，以及鲁棒性，在面对不可分离的数据。
We present AROW, an online learning algorithm for binary and multiclass problems that combines large margin training, confidence weighting, and the capacity to handle non-separable data. AROW performs adaptive regularization of the prediction function upon seeing each new instance, allowing it to perform especially well in the presence of label noise. We derive mistake bounds for the binary and multiclass settings that are similar in form to the second order perceptron bound. Our bounds do not assume separability. We also relate our algorithm to recent confidence-weighted online learning techniques. Empirical evaluations show that AROW achieves state-of-the-art performance on a wide range of binary and multiclass tasks, as well as robustness in the face of non-separable data.