{"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-generate-with-memory","title":"Learning to Generate with Memory","arxiv_id":"1602.07416","date":"2016-02-24","proceeding":null,"authors":["Chongxuan Li","Jun Zhu","Bo Zhang"],"abstract":"Memory units have been widely used to enrich the capabilities of deep\nnetworks on capturing long-term dependencies in reasoning and prediction tasks,\nbut little investigation exists on deep generative models (DGMs) which are good\nat inferring high-level invariant representations from unlabeled data. This\npaper presents a deep generative model with a possibly large external memory\nand an attention mechanism to capture the local detail information that is\noften lost in the bottom-up abstraction process in representation learning. By\nadopting a smooth attention model, the whole network is trained end-to-end by\noptimizing a variational bound of data likelihood via auto-encoding variational\nBayesian methods, where an asymmetric recognition network is learnt jointly to\ninfer high-level invariant representations. The asymmetric architecture can\nreduce the competition between bottom-up invariant feature extraction and\ntop-down generation of instance details. Our experiments on several datasets\ndemonstrate that memory can significantly boost the performance of DGMs and\neven achieve state-of-the-art results on various tasks, including density\nestimation, image generation, and missing value imputation.","url_abs":"http://arxiv.org/abs/1602.07416v2","url_pdf":"http://arxiv.org/pdf/1602.07416v2.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-generate-with-memory","repo_url":"https://github.com/zhenxuan00/MEM_DGM","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"density-estimation","task_name":"Density Estimation"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"imputation","task_name":"Imputation"},{"task_slug":"representation-learning","task_name":"Representation Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1602.07416","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}