{"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/learning-to-bid-without-knowing-your-value","title":"Learning to Bid Without Knowing your Value","arxiv_id":"1711.01333","date":"2017-11-03","proceeding":null,"authors":["Zhe Feng","Chara Podimata","Vasilis Syrgkanis"],"abstract":"We address online learning in complex auction settings, such as sponsored\nsearch auctions, where the value of the bidder is unknown to her, evolving in\nan arbitrary manner and observed only if the bidder wins an allocation. We\nleverage the structure of the utility of the bidder and the partial feedback\nthat bidders typically receive in auctions, in order to provide algorithms with\nregret rates against the best fixed bid in hindsight, that are exponentially\nfaster in convergence in terms of dependence on the action space, than what\nwould have been derived by applying a generic bandit algorithm and almost\nequivalent to what would have been achieved in the full information setting.\nOur results are enabled by analyzing a new online learning setting with\noutcome-based feedback, which generalizes learning with feedback graphs. We\nprovide an online learning algorithm for this setting, of independent interest,\nwith regret that grows only logarithmically with the number of actions and\nlinearly only in the number of potential outcomes (the latter being very small\nin most auction settings). Last but not least, we show that our algorithm\noutperforms the bandit approach experimentally and that this performance is\nrobust to dropping some of our theoretical assumptions or introducing noise in\nthe feedback that the bidder receives.","url_abs":"http://arxiv.org/abs/1711.01333v5","url_pdf":"http://arxiv.org/pdf/1711.01333v5.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":"learning-to-bid-without-knowing-your-value","repo_url":"https://github.com/zfengharvard/bandit-sponsored-search","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.01333","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}