{"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/attend-before-you-act-leveraging-human-visual","title":"Attend Before you Act: Leveraging human visual attention for continual learning","arxiv_id":"1807.09664","date":"2018-07-25","proceeding":null,"authors":["Khimya Khetarpal","Doina Precup"],"abstract":"When humans perform a task, such as playing a game, they selectively pay\nattention to certain parts of the visual input, gathering relevant information\nand sequentially combining it to build a representation from the sensory data.\nIn this work, we explore leveraging where humans look in an image as an\nimplicit indication of what is salient for decision making. We build on top of\nthe UNREAL architecture in DeepMind Lab's 3D navigation maze environment. We\ntrain the agent both with original images and foveated images, which were\ngenerated by overlaying the original images with saliency maps generated using\na real-time spectral residual technique. We investigate the effectiveness of\nthis approach in transfer learning by measuring performance in the context of\nnoise in the environment.","url_abs":"http://arxiv.org/abs/1807.09664v1","url_pdf":"http://arxiv.org/pdf/1807.09664v1.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":"attend-before-you-act-leveraging-human-visual","repo_url":"https://github.com/kkhetarpal/unrealwithattention","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"continual-learning","task_name":"Continual Learning"},{"task_slug":"decision-making","task_name":"Decision Making"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1807.09664","atlas_url":"https://app.syntology.ai/?focus=1807.09664","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}