Exploiting Modern Hardware for High-Dimensional Nearest Neighbor Search
This work addresses the need for faster nearest neighbor search in large-scale databases, but it is incremental as it builds on existing product quantization methods.
The thesis tackles the problem of efficient high-dimensional nearest neighbor search for large-scale multimedia and machine learning applications by building on product quantization, and it proposes contributions that exploit modern CPU capabilities like SIMD and cache hierarchy to further decrease response times.
Many multimedia information retrieval or machine learning problems require efficient high-dimensional nearest neighbor search techniques. For instance, multimedia objects (images, music or videos) can be represented by high-dimensional feature vectors. Finding two similar multimedia objects then comes down to finding two objects that have similar feature vectors. In the current context of mass use of social networks, large scale multimedia databases or large scale machine learning applications are more and more common, calling for efficient nearest neighbor search approaches. This thesis builds on product quantization, an efficient nearest neighbor search technique that compresses high-dimensional vectors into short codes. This makes it possible to store very large databases entirely in RAM, enabling low response times. We propose several contributions that exploit the capabilities of modern CPUs, especially SIMD and the cache hierarchy, to further decrease response times offered by product quantization.