[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85688-en":3,"doc-seo-85688-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},85688,137441390410,"Hazel","https://ap-avatar.wpscdn.com/avatar/2000252f4ab5702993?_k=1776741390130283984",8,"Research & Report","Learning High-Level Decision Making with an Interaction-Aware Attention-Based Network in Autonomous Driving","Reliable learning-based high-level decision making for lane changes and speed control in automated driving must handle dynamically sized inputs as traffic patterns shift. Existing shared-encoder methods based on DeepSet variants achieve scalable aggregation but omit explicit traffic interaction modeling, reducing performance in negotiation-rich scenes such as intersections. Attention-based approaches model interactions yet suffer quadratic cost and coarse representation control. DecisionPerceiver projects variable agents into a fixed latent space to regulate granularity via latent queries. Evaluations across multiple scenarios show consistent gains and scalability as vehicle counts increase.","Learning High-Level Decision Making with an Interaction-Aware Attention-Based Network in Autonomous Driving  \nMarcelo Contreras 1 , Willi Poh 2 , Christoph Stiller 2 , Ehsan Hashemi 1 ,  \narXiv :2607 .09725v1 [ cs .RO] 29 Jun 2026  \nAbstract—Reliable learning-based high-level decision making for lane changes and speed control in automated driving must accommodate dynamically sized inputs due to varying scene traffic flow. DeepSet and its variants represent the state of the art among shared-encoder approaches; however, they neglect explicit traffic interaction modeling, limiting performance in negotiationintensive scenarios such as intersections. Attention-based methods capture interactions among static and dynamic agents, but incur quadratic memory and computational complexity and provide limited control over representation granularity. Inspired by Perceiver IO, an attention-based architecture, DecisionPerceiver, is proposed to project dynamic agent features into a fixedsize latent space, where feature granularity is regulated by the number of latent queries, improving scalability for larger networks. A finer discretization of the action set is further proposed to increase the performance gain due to interaction awareness. Extensive evaluations across three driving scenarios that require different levels of interaction awareness demonstrate consistent performance gains and generalization across various navigation objectives. In addition, the proposed architecture is assessed in scenarios with an increasing number of vehicles to demonstrate scalability.  \nI. INTRODUCTION  \nReinforcement learning (RL) has emerged as a powerful framework for autonomous driving policy learning [1]–[3], as it optimizes decision-making through closed-loop interaction with the environment. In contrast to imitation learning (IL), which is limited by the distribution and coverage of expert demonstrations [4], RL enables performance beyond demonstrated behaviors and improved generalization to novel scenarios. A key challenge in neural policy representations is sensitivity to input ordering, as the number and arrangement of surrounding agents vary dynamically with perception outputs. Prior works address this using convolutional neural networks (CNNs) with grid-based or bird’s-eye view (BEV) representations centered one the ego vehicle [5] . While effective for spatial encoding, such inputs introduce redundancy and may obscure explicit kinematic states critical for decision making. To address permutation sensitivity, the DeepSets architecture [6] was introduced to process variable-size inputs using shared feature encoders combined with permutation-invariant aggregation operators, such as pooling or summation. This formulation has been extended for long-horizon reasoning [7], surrogate-objective training [8], explainable policies [9], and  \n1 M. Contreras and E. Hashemi (Corresponding Author, [ehashemi@ualberta.ca](ehashemi@ualberta.ca)) are with the NODE lab, University of Alberta, Edmonton, AB, T6G 1H9 Canada.  \n2 W. Poh and C. Stiller are with the MRT Institute, Karlsruhe Institute of Technology, Karlsruhe, Germany.  \nFig. 1: Traffic-participant interactions are modeled by projecting agent features into a latent space. Based on agents’kinematic states and inferred interaction strength, the ego vehicle selects a full left-lane-change maneuver. Blue edges depict ego-to-agent interactions, with thickness proportional to intensity. The discrete action set uses finer lane-change and speed-adjustment discretization to improve trajectory tracking.  \nmitigation of multilayer perceptrons’ spectral bias via Fourier feature mappings [10] .  \nDeepSets exhibits a key limitation: feature collapse after aggregating shared encoder outputs into a single vector, which restricts explicit modeling of inter-agent interactions essential for negotiation scenarios. To capture such interactions, graph neural networks (GNNs) [10]–[12] model vehicles as nodes and interaction stre","cbCail5nGcFOt6ih","https://ap.wps.com/l/cbCail5nGcFOt6ih","pdf",565991,1,6,"English","en",105,"# Introduction\n## Key challenge in neural policy representations\n## From DeepSets to interaction modeling\n## Limits of GNNs and attention complexity\n# Proposed approach: DecisionPerceiver","[{\"question\":\"What problem does the document address in learning-based decision making for autonomous driving?\",\"answer\":\"It addresses reliable lane-change and speed-control learning when input size varies with traffic flow, which makes interaction modeling and scalable representations difficult.\"},{\"question\":\"Why do DeepSet-style shared-encoder approaches fall short?\",\"answer\":\"They aggregate agent features into a single vector and can experience feature collapse, which limits explicit inter-agent interaction modeling needed for negotiation scenarios.\"},{\"question\":\"How does DecisionPerceiver improve efficiency while preserving interaction awareness?\",\"answer\":\"It projects variable-size agent features into a fixed-size latent space inspired by Perceiver IO, so computational cost is decoupled from input size and representation granularity is controlled by latent queries.\"}]",1784205612,15,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"learning-high-level-decision-making-with-an-interaction-aware-attention-based-network-in-autonomous-driving","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/learning-high-level-decision-making-with-an-interaction-aware-attention-based-network-in-autonomous-driving/85688/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the document address in learning-based decision making for autonomous driving?","Question",{"text":75,"@type":76},"It addresses reliable lane-change and speed-control learning when input size varies with traffic flow, which makes interaction modeling and scalable representations difficult.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why do DeepSet-style shared-encoder approaches fall short?",{"text":80,"@type":76},"They aggregate agent features into a single vector and can experience feature collapse, which limits explicit inter-agent interaction modeling needed for negotiation scenarios.",{"name":82,"@type":73,"acceptedAnswer":83},"How does DecisionPerceiver improve efficiency while preserving interaction awareness?",{"text":84,"@type":76},"It projects variable-size agent features into a fixed-size latent space inspired by Perceiver IO, so computational cost is decoupled from input size and representation granularity is controlled by latent queries.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]