Neuromorphic Nearest-Neighbor Search Using Intel's Pohoiki Springs
This work addresses the challenge of efficient and low-energy database search for applications like machine learning and data retrieval, representing an incremental improvement by applying neuromorphic computing to an existing problem.
The authors tackled the problem of nearest-neighbor search in large databases by developing a scalable approximate k-NN algorithm using Intel's Pohoiki Springs neuromorphic system, achieving superior latency, index build time, and energy efficiency compared to state-of-the-art CPU-based implementations on datasets with over 1 million high-dimensional patterns.
Neuromorphic computing applies insights from neuroscience to uncover innovations in computing technology. In the brain, billions of interconnected neurons perform rapid computations at extremely low energy levels by leveraging properties that are foreign to conventional computing systems, such as temporal spiking codes and finely parallelized processing units integrating both memory and computation. Here, we showcase the Pohoiki Springs neuromorphic system, a mesh of 768 interconnected Loihi chips that collectively implement 100 million spiking neurons in silicon. We demonstrate a scalable approximate k-nearest neighbor (k-NN) algorithm for searching large databases that exploits neuromorphic principles. Compared to state-of-the-art conventional CPU-based implementations, we achieve superior latency, index build time, and energy efficiency when evaluated on several standard datasets containing over 1 million high-dimensional patterns. Further, the system supports adding new data points to the indexed database online in O(1) time unlike all but brute force conventional k-NN implementations.