{"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/accurate-and-diverse-sampling-of-sequences","title":"Accurate and Diverse Sampling of Sequences based on a \"Best of Many\" Sample Objective","arxiv_id":"1806.07772","date":"2018-06-20","proceeding":null,"authors":["Apratim Bhattacharyya","Bernt Schiele","Mario Fritz"],"abstract":"For autonomous agents to successfully operate in the real world, anticipation\nof future events and states of their environment is a key competence. This\nproblem has been formalized as a sequence extrapolation problem, where a number\nof observations are used to predict the sequence into the future. Real-world\nscenarios demand a model of uncertainty of such predictions, as predictions\nbecome increasingly uncertain -- in particular on long time horizons. While\nimpressive results have been shown on point estimates, scenarios that induce\nmulti-modal distributions over future sequences remain challenging. Our work\naddresses these challenges in a Gaussian Latent Variable model for sequence\nprediction. Our core contribution is a \"Best of Many\" sample objective that\nleads to more accurate and more diverse predictions that better capture the\ntrue variations in real-world sequence data. Beyond our analysis of improved\nmodel fit, our models also empirically outperform prior work on three diverse\ntasks ranging from traffic scenes to weather data.","url_abs":"http://arxiv.org/abs/1806.07772v2","url_pdf":"http://arxiv.org/pdf/1806.07772v2.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":"accurate-and-diverse-sampling-of-sequences","repo_url":"https://github.com/apratimbhattacharyya18/CGM_BestOfMany","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"human-pose-forecasting","task_name":"Human Pose Forecasting"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/human-pose-forecasting-on-human36m","task":"Human Pose Forecasting","dataset":"Human3.6M","model":"BoM","rank_in_archive_order":31,"of":33,"metrics":{"ADE":"448","APD":"6265","FDE":"533","MMADE":"514","MMFDE":"544"},"uses_additional_data":false},{"leaderboard":"/sota/human-pose-forecasting-on-humaneva-i","task":"Human Pose Forecasting","dataset":"HumanEva-I","model":"BoM","rank_in_archive_order":6,"of":11,"metrics":{"ADE@2000ms":"271","APD@2000ms":"2846","FDE@2000ms":"279"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1806.07772","atlas_url":"https://app.syntology.ai/?focus=1806.07772","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}