Current browse context:
cs.LG
Change to browse by:
References & Citations
Statistics > Machine Learning
Title: Stream Clipper: Scalable Submodular Maximization on Stream
(Submitted on 1 Jun 2016 (v1), last revised 13 Feb 2018 (this version, v3))
Abstract: We propose a streaming submodular maximization algorithm "stream clipper" that performs as well as the offline greedy algorithm on document/video summarization in practice. It adds elements from a stream either to a solution set $S$ or to an extra buffer $B$ based on two adaptive thresholds, and improves $S$ by a final greedy step that starts from $S$ adding elements from $B$. During this process, swapping elements out of $S$ can occur if doing so yields improvements. The thresholds adapt based on if current memory utilization exceeds a budget, e.g., it increases the lower threshold, and removes from the buffer $B$ elements below the new lower threshold. We show that, while our approximation factor in the worst case is $1/2$ (like in previous work, and corresponding to the tight bound), we show that there are data-dependent conditions where our bound falls within the range $[1/2, 1-1/e]$. In news and video summarization experiments, the algorithm consistently outperforms other streaming methods, and, while using significantly less computation and memory, performs similarly to the offline greedy algorithm.
Submission history
From: Tianyi Zhou [view email][v1] Wed, 1 Jun 2016 18:43:13 GMT (4422kb,D)
[v2] Sat, 10 Feb 2018 22:36:44 GMT (4455kb,D)
[v3] Tue, 13 Feb 2018 01:50:38 GMT (4456kb,D)
Link back to: arXiv, form interface, contact.