New Loss Functions for Fast Maximum Inner Product Search
Year: 2019
Type: preprint
Abstract: Quantization based methods are popular for solving large scale maximum innerproduct search problems. However, in most traditional quantization works, the
objective is to minimize the reconstruction error for datapoints to be
searched. In this work, we focus directly on minimizing error in inner product
approximation and derive a new class of quantization loss functions. One key
aspect of the new loss functions is that we weight the error term based on the
value of the inner product, giving more importance to pairs of queries and
datapoints whose inner products are high. We provide theoretical grounding to
the new quantization loss function, which is simple, intuitive and able to work
with a variety of quantization techniques, including binary quantization and
product quantization. We conduct experiments on standard benchmarking datasets
to demonstrate that our method using the new objective outperforms other
state-of-the-art methods.
Source: arXiv (Cornell University)
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2971288826