Ternary and Binary Quantization for Improved Classification
AbstractDimension 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 extreme quantization will
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 preserved 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.