[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-142570-105":59,"doc-detail-142570-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","entity-abstraction-in-visual-model-based-reinforcement-learning-read-abstract","Entity Abstraction in Visual Model-Based Reinforcement Learning - read abstract","","The paper introduces OP3, a model-based reinforcement learning framework that learns object/entity representations from raw visual observations without supervision and then uses them for prediction and planning. An interactive inference algorithm grounds abstract entity representations to real world objects by incorporating dynamic scene information. A symmetric, locally scoped entity encoder handles a variable number of entities. On block-stacking tasks, OP3 generalizes to novel configurations and more objects than in training, outperforming object-supervised baselines and a state-of-the-art video prediction model.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/entity-abstraction-in-visual-model-based-reinforcement-learning-read-abstract/142570/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/entity-abstraction-in-visual-model-based-reinforcement-learning-read-abstract/142570.png","ImageObject",300,407,{"name":92,"@type":93},"วิน","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-17","2026-08-25",true,{"@type":102,"interactionType":103,"userInteractionCount":34},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What is OP3 and what does it learn from visual input?","Question",{"text":112,"@type":113},"OP3 is a model-based reinforcement learning framework that acquires entity/object representations directly from raw visual observations without supervision, and then uses these representations to predict and plan.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does OP3 connect abstract entity variables to real objects in the world?",{"text":117,"@type":113},"It uses an interactive inference algorithm that incorporates dynamic information in the scene to infer how abstract entity variables correspond to concrete objects.",{"name":119,"@type":110,"acceptedAnswer":120},"How does OP3 scale to scenes with a variable number of entities?",{"text":121,"@type":113},"OP3 processes each inferred entity representation symmetrically using the same locally scoped function, allowing the model to handle different counts of entities.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},142570,1787683574,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":34,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":129,"read_time":144,"preview_page_count":4},2336475104736,"https://ap-avatar.wpscdn.com/avatar/22000c4c5e0e5b17e70?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786591360781797222","Entity Abstraction in Visual Model-Based Reinforcement Learning  \nRishi Veerapaneni􀀃 ; 1 , John D. Co-Reyes􀀃 ; 1 , Michael Chang􀀃 ; 1 , Michael Janner 1 Chelsea Finn2 , Jiajun Wu3 , Joshua Tenenbaum3 , Sergey Levine 1  \nAbstract: We present OP3, a framework for model-based reinforcement learning that acquires object representations from raw visual observations without supervision and uses them to predict and plan. To ground these abstract representations of entities to actual objects in the world, we formulate an interactive inference algorithm which incorporates dynamic information in the scene. Our model can handle a variable number of entities by symmetrically processing each object representation with the same locally-scoped function. On block-stacking tasks, OP3 can generalize to novel block configurations and more objects than seen during training, outperforming both a model that assumes access to object supervision anda state-of-the-art video prediction model.  \nKeywords: model-based reinforcement learning, objects, compositionality  \n1 Introduction  \nA powerful tool for modeling the complexity of the physical world is to frame this complexity as the composition of simpler entities and processes. For example, the study of classical mechanics in terms of macroscopic objects and a small set of laws governing their motion has enabled not only an explanation of natural phenomena like apples falling from trees but the invention of structures that never before existed in human history, such as skyscrapers. Paradoxically, the creative variation of such physical constructions in human society is due in part to the uniformity with which human models of physical laws apply to the literal building blocks that comprise such structures – the reuse of the same simpler models that apply to primitive entities and their relations in different ways obviates the need, and cost, of designing custom solutions from scratch for each construction instance.  \nThe challenge of scaling the generalization abilities of learning robots follow a similar characteristic to the challenges of modeling physical phenomena: the complexity of the task space may scale combinatorially with the configurations and number  \nFigure 1: OP3 infers a set of entity vari  \nables H1(t:K) from an observation xt and predicts their future states given a sequence of actions a (t:T ) . We evaluate rollouts during planning by scoring these predicted states against inferred goal hidden states hk(G) .  \nof objects, but if all scene instances share the same set of objects that follow the same physical laws, then transforming the problem of modeling scenes into a problem of modeling objects and the local physical processes that govern their interactions may provide a significant benefit in generalizing to solving novel physical tasks the learner has not encountered before.  \nThat is the central hypothesis of this paper, which we test by defining models for perceiving and predicting raw observations that are themselves compositions of simpler functions that operate locally on entities rather than globally on scenes. Importantly, the symmetry that all objects follow the same physical laws enables us to define these learnable entity-centric functions to take as input argument  \n* Equal contribution.  \n1University of California Berkeley. 2 Stanford University. 3Massachusetts Institute of Technology.  \n3rd Conference on Robot Learning (CoRL 2019), Osaka, Japan.  \nFigure 2: Our approach to model-based RL imposes the entity abstraction: (a) The hidden state is factorized into local entity states, each of which are symmetrically processed with the same function that takes in a generic entity as an argument. In contrast, prior work do not incorporate symmetric processing of the hidden state. (b) The hidden state represents the scene globally and is processed with a single function. (c) Different chunks of the hidden state represent different entities and the entire state is processed with a sin","cbCaimRwU0UaDy7O","https://ap.wps.com/l/cbCaimRwU0UaDy7O","pdf",2995857,18,"English","# Introduction\n## Entity abstraction hypothesis\n## Object-centric perception, prediction, and planning (OP3)\n## Interactive inference for grounding\n## Empirical evaluation on block-stacking tasks","[{\"question\":\"What is OP3 and what does it learn from visual input?\",\"answer\":\"OP3 is a model-based reinforcement learning framework that acquires entity/object representations directly from raw visual observations without supervision, and then uses these representations to predict and plan.\"},{\"question\":\"How does OP3 connect abstract entity variables to real objects in the world?\",\"answer\":\"It uses an interactive inference algorithm that incorporates dynamic information in the scene to infer how abstract entity variables correspond to concrete objects.\"},{\"question\":\"How does OP3 scale to scenes with a variable number of entities?\",\"answer\":\"OP3 processes each inferred entity representation symmetrically using the same locally scoped function, allowing the model to handle different counts of entities.\"}]","Entity Abstraction in Visual Model-Based Reinforcement Learning - read abstract | PDF",45]