{"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/scalable-metric-learning-via-weighted","title":"Scalable Metric Learning via Weighted Approximate Rank Component Analysis","arxiv_id":"1603.00370","date":"2016-03-01","proceeding":null,"authors":["Cijo Jose","Francois Fleuret"],"abstract":"We are interested in the large-scale learning of Mahalanobis distances, with\na particular focus on person re-identification.\n  We propose a metric learning formulation called Weighted Approximate Rank\nComponent Analysis (WARCA). WARCA optimizes the precision at top ranks by\ncombining the WARP loss with a regularizer that favors orthonormal linear\nmappings, and avoids rank-deficient embeddings. Using this new regularizer\nallows us to adapt the large-scale WSABIE procedure and to leverage the Adam\nstochastic optimization algorithm, which results in an algorithm that scales\ngracefully to very large data-sets. Also, we derive a kernelized version which\nallows to take advantage of state-of-the-art features for re-identification\nwhen data-set size permits kernel computation.\n  Benchmarks on recent and standard re-identification data-sets show that our\nmethod beats existing state-of-the-art techniques both in term of accuracy and\nspeed. We also provide experimental analysis to shade lights on the properties\nof the regularizer we use, and how it improves performance.","url_abs":"http://arxiv.org/abs/1603.00370v2","url_pdf":"http://arxiv.org/pdf/1603.00370v2.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":[],"tasks":[{"task_slug":"metric-learning","task_name":"Metric Learning"},{"task_slug":"person-re-identification","task_name":"Person Re-Identification"},{"task_slug":"stochastic-optimization","task_name":"Stochastic Optimization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/person-re-identification-on-market-1501","task":"Person Re-Identification","dataset":"Market-1501","model":"WARCA","rank_in_archive_order":122,"of":135,"metrics":{"Rank-1":"45.16"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}