PANENE: A Progressive Algorithm for Indexing and Querying Approximate k-Nearest Neighbors

Abstract

We present PANENE, a progressive algorithm for approximate nearest neighbor indexing and querying. Although the use of k-nearest neighbor (KNN) libraries is common in many data analysis methods, most KNN algorithms can only be queried when the whole dataset has been indexed, i.e., they are not online. Even the few online implementations are not progressive in the sense that the time to index incoming data is not bounded and cannot satisfy the latency requirements of progressive systems. This long latency has significantly limited the use of many machine learning methods, such as t-SNE, in interactive visual analytics. PANENE is a novel algorithm for Progressive Approximate k-NEarest NEighbors, enabling fast KNN queries while continuously indexing new batches of data. Following the progressive computation paradigm, PANENE operations can be bounded in time, allowing analysts to access running results within an interactive latency. PANENE can also incrementally build and maintain a cache data structure, a KNN lookup table, to enable constant-time lookups for KNN queries. Finally, we present three progressive applications of PANENE, such as regression, density estimation, and responsive t-SNE, opening up new opportunities to use complex algorithms in interactive systems.

Pyramid-based Scatterplots Sampling for Progressive and Streaming Data Visualization

“…Several progressive systems and algorithms have been designed recently using high-dimensional projections visualized with density maps [20,25,27,34,44]. However, to the best of our knowledge, there are no progressive sampling approaches designed for scatterplots that satisfy the requirements of progressive and streaming visualization, and our approach fills this gap.…”

“…Competitive analysis [7] is a widely-used approach for evaluating online algorithms, by comparing the performance with an equivalent offline algorithm. In our progressive setting, users will rely on intermediate results to make early decisions and thus we measure the performance of each frame, following the practice in Jo et al [25].…”

ProReveal: Progressive Visual Analytics With Safeguards

“…Pezzotti et al [23] introduced approximate t-Distributed Stochastic Neighbor Embedding (A-tSNE), which significantly lowered the long latency of the t-SNE algorithm [24]. Jo et al [25] focused on further reducing the initial delay of t-SNE through progressive neighbor computation and presented a responsive t-SNE algorithm. Turkay et al [26] proposed DimXplorer that integrates the incremental PCA and mini-batch k-means clustering algorithms with data exploration.…”

Real-time currency recognition on video using AKAZE algorithm

“…Therefore, matching features with K-NN is considered inefficient, so the FLANN method for matching multi-dimensional data is needed. The FLANN method uses the K-Dimensional Tree (KD-Tree) to represent multi-dimensional binary tree data to separate certain areas based on their value position [20].…”