{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/scaling-active-search-using-linear-similarity","title":"Scaling Active Search using Linear Similarity Functions","arxiv_id":"1705.00334","date":"2017-04-30","proceeding":null,"authors":["Sibi Venkatesan","James K. Miller","Jeff Schneider","Artur Dubrawski"],"abstract":"Active Search has become an increasingly useful tool in information retrieval\nproblems where the goal is to discover as many target elements as possible\nusing only limited label queries. With the advent of big data, there is a\ngrowing emphasis on the scalability of such techniques to handle very large and\nvery complex datasets.\n  In this paper, we consider the problem of Active Search where we are given a\nsimilarity function between data points. We look at an algorithm introduced by\nWang et al. [2013] for Active Search over graphs and propose crucial\nmodifications which allow it to scale significantly. Their approach selects\npoints by minimizing an energy function over the graph induced by the\nsimilarity function on the data. Our modifications require the similarity\nfunction to be a dot-product between feature vectors of data points, equivalent\nto having a linear kernel for the adjacency matrix. With this, we are able to\nscale tremendously: for $n$ data points, the original algorithm runs in\n$O(n^2)$ time per iteration while ours runs in only $O(nr + r^2)$ given\n$r$-dimensional features.\n  We also describe a simple alternate approach using a weighted-neighbor\npredictor which also scales well. In our experiments, we show that our method\nis competitive with existing semi-supervised approaches. We also briefly\ndiscuss conditions under which our algorithm performs well.","url_abs":"http://arxiv.org/abs/1705.00334v2","url_pdf":"http://arxiv.org/pdf/1705.00334v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"scaling-active-search-using-linear-similarity","repo_url":"https://github.com/bkj/simple_las","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"information-retrieval","task_name":"Information Retrieval"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}