Scaling Graph-based Semi Supervised Learning to Large Number of Labels Using Count-Min Sketch
This addresses a critical bottleneck for applications with very large label sets, such as in big data domains, by enabling scaling to one million labels, which was previously infeasible.
The paper tackles the scalability issue of graph-based semi-supervised learning algorithms, which have high space and time complexity with large numbers of labels, by proposing MAD-SKETCH, a method that uses Count-min Sketch to reduce space from O(m) to O(log m) and achieves up to 10x speedup while maintaining similar performance.
Graph-based Semi-supervised learning (SSL) algorithms have been successfully used in a large number of applications. These methods classify initially unlabeled nodes by propagating label information over the structure of graph starting from seed nodes. Graph-based SSL algorithms usually scale linearly with the number of distinct labels (m), and require O(m) space on each node. Unfortunately, there exist many applications of practical significance with very large m over large graphs, demanding better space and time complexity. In this paper, we propose MAD-SKETCH, a novel graph-based SSL algorithm which compactly stores label distribution on each node using Count-min Sketch, a randomized data structure. We present theoretical analysis showing that under mild conditions, MAD-SKETCH can reduce space complexity at each node from O(m) to O(log m), and achieve similar savings in time complexity as well. We support our analysis through experiments on multiple real world datasets. We observe that MAD-SKETCH achieves similar performance as existing state-of-the-art graph- based SSL algorithms, while requiring smaller memory footprint and at the same time achieving up to 10x speedup. We find that MAD-SKETCH is able to scale to datasets with one million labels, which is beyond the scope of existing graph- based SSL algorithms.