Results of the Big ANN: NeurIPS’23 competition

Kai Zhang (The Ohio State University) · Harsha Vardhan simhadri (Microsoft ) · Martin Aumüller (IT University of Copenhagen) · Matthijs Douze (Facebook AI Research) · Dmitry Baranchuk (Yandex) · Amir Ingber (Pinecone) · Edo Liberty (Yale University) · George Williams (Ansible AI) · Ben Landrum (Cornell University) · Magdalen Manohar (Carnegie Mellon University) · Mazin Karjikar (University of Maryland, College Park) · Laxman Dhulipala (UMD) · Meng Chen (Fudan University) · Yue Chen (Fudan University) · Rui Ma (Jilin University) · Yuzheng Cai (Fudan University) · Jiayang Shi (Fudan University) · Weiguo Zheng (Fudan University) · Yizhuo Chen (Fudan University) · Jie Yin (Tencent) · Ben Huang (Baidu)
advancements in aiapproximate nearest neighborcompetition trackscomputational resourcesefficiency improvementsevaluation metricsfiltered variantsindexing data structuresinnovative solutionsout-of-distributionsearch accuracysearch algorithmssparse variantsstandard datasetsstreaming variants

The 2023 Big ANN Challenge, held at NeurIPS 2023, focused on advancing the state-of-the-art in indexing data structures and search algorithms for practical variants of Approximate Nearest Neighbor (ANN) search that reflect its the growing complexity and diversity of workloads. Unlike prior challenges that emphasized scaling up classical ANN search (Simhadri et al., NeurIPS 2021), this competition addressed sparse, filtered, out-of-distribution, and streaming variants of ANNS. Participants developed and submitted innovative solutions that were evaluated on new standard datasets with constrained computational resources. The results showcased significant improvements in search accuracy and efficiency, with notable contributions from both academic and industrial teams. This paper summarizes the competition tracks, datasets, evaluation metrics, and the innovative approaches of the top-performing submissions, providing insights into the current advancements and future directions in the field of approximate nearest neighbor search.