We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: Stream Clipper: Scalable Submodular Maximization on Stream

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.
Comments: 17 pages, 12 figures, submitted to conference
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Combinatorics (math.CO)
Cite as: arXiv:1606.00389 [stat.ML]
  (or arXiv:1606.00389v3 [stat.ML] for this version)

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.