论文标题
用于改进分类的三元和二进制量化
Ternary and Binary Quantization for Improved Classification
论文作者
论文摘要
降低和数据量化是降低数据复杂性的两种重要方法。在本文中，我们研究了首先通过随机投影降低数据维度的方法，然后将投影量化为三元或二元代码，这些预测已广泛应用于分类。通常，由于高量化错误，量化将严重降低分类的准确性。然而，有趣的是，我们观察到量化可以提供可比的且通常是优越的准确性，因为要量化的数据是使用常见过滤器产生的稀疏特征。此外，如果特征和随机投影矩阵都足够稀疏，则可以将此量化属性保持在稀疏特征的随机投影中。通过进行广泛的实验，我们验证和分析了这一有趣的特性。
Dimension reduction and data quantization are two important methods for reducing data complexity. In the paper, we study the methodology of first reducing data dimension by random projection and then quantizing the projections to ternary or binary codes, which has been widely applied in classification. Usually, the quantization will seriously degrade the accuracy of classification due to high quantization errors. Interestingly, however, we observe that the quantization could provide comparable and often superior accuracy, as the data to be quantized are sparse features generated with common filters. Furthermore, this quantization property could be maintained in the random projections of sparse features, if both the features and random projection matrices are sufficiently sparse. By conducting extensive experiments, we validate and analyze this intriguing property.
