[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-116869-en":3,"doc-seo-116869-105":29,"detail-sidebar-cat-0-en-105":90},{"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":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":11,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":26,"seo_description":14,"update_tm":27,"read_time":28},116869,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253",8,"Research & Report","LOBIN - In-Network Machine Learning for Limit Order Books","Machine learning is reshaping algorithmic trading, yet fast execution latency remains a decisive constraint. The dual objective of stronger ML capability and lower trading delays is difficult to achieve simultaneously, motivating a new computing placement strategy. LOBIN delivers machine learning market prediction by constructing limit order books from high-frequency feed data and performing inference inside programmable switches. Compared with server-based approaches, LOBIN attains lower latency, higher throughput, and only minor degradation in prediction performance while using microstructure signals for time-series forecasting.","LOBIN: In-Network Machine Learning for Limit Order Books  \nXinpeng Hong, Changgang Zheng, Stefan Zohren, and Noa Zilberman  \nDepartment of Engineering Science, University of Oxford, Oxford, United Kingdom {xinpeng.hong, changgang.zheng, stefan.zohren, [noa.zilberman](noa.zilberman}@eng.ox.ac.uk)[}](noa.zilberman}@eng.ox.ac.uk)[@eng.ox.ac.uk](noa.zilberman}@eng.ox.ac.uk)  \nAbstract—Machine learning is driving the evolution of algorithmic trading, but the demands for fast execution speed remain. Although both aim to increase profitability, embedding more powerful machine learning approaches and lowering trading latencies are hard to achieve simultaneously. Offloading machine learning inference to programmable network devices, also referred to as in-network machine learning, provides a delicate balance between the two ends of this trade-off. In this paper, we present LOBIN, providing machine learning based market prediction using high-frequency market data feeds. LOBIN builds limit order books and conducts inference within programmable switches. Compared with server-based solutions, LOBIN predicts future stock price movements with lower latency, higher throughput, and a minor impact on machine learning performance.  \nIndex Terms—In-network computing, machine learning, programmable switches, P4, microstructure market data, limit order books, time series prediction.  \nI. INTRODUCTION  \nAlgorithmic trading has been growing over the past few decades as financial firms automate processes traditionally done by human traders. It uses computer algorithms to automatically execute orders under preset trading instructions [1] . As an essential form of algorithmic trading, high-frequency trading (HFT) is characterized by placing larger numbers of orders within a minimum time and being able to react quickly under changing market conditions [2] . The latency of the HFT market participants has been measured in microseconds [3] .  \nThe rise of artificial intelligence further drives the growth of high-frequency algorithmic trading, with machine learning (ML) approaches becoming widespread in the field of HFT [4],[5] . However, the increasing complexity of ML models used also challenges existing trading systems, creating a demand for latency reduction throughout the trading process.  \nA common HFT problem is predicting future price movements from market microstructure signals, which has been proven to be feasible and profitable [4] . There are a number of previous works focusing on market forecasting utilizing electronic limit order books (LOBs) combined with ML models [6]–[9] . For a particular stock, a real-time LOB is constructed from unmatched limit orders that are predetermined with specific prices. It contains a wealth of information that can be used as ML features [10] .  \nThis work was partly funded by VMware and we acknowledge support from Intel. Stefan Zohren thanks the Oxford-Man Institute of Quantitative Finance for financial support. We also thank Zihao Zhang for the valuable discussions on this work.  \nFig. 1. General working scenario of LOBIN.  \nIn-network computing offloads applications to run into programmable network devices [11] . In-network ML, as a specific type of in-network computing, deploys pre-trained ML models within network devices and conducts inference therefor lower latency, higher throughput, and more efficient power utilization [12], [13] . By design, in-network ML provides a practical solution for reducing the latency of time-sensitive financial applications, such as trading scenarios.  \nAs nearly no previous works focus on time-sensitive applications of in-network ML, this paper provides an in-depth study on the application of in-network ML to market prediction using LOBs. We design and implement LOBIN, an innetwork prototype for future price movement prediction by maintaining a LOB based on market-by-order (MBO) data feeds. LOBIN stands for “Limit Order Books In Network”. Figure 1 shows a general working scenari","cbCaiifLtVAju6MM","https://ap.wps.com/l/cbCaiifLtVAju6MM","pdf",1760523,1,"English","en",105,"# Introduction\n## In-network computing and in-network ML\n# Background and Motivation\n## ML for Market Prediction","[{\"question\":\"What problem does LOBIN address in algorithmic trading?\",\"answer\":\"LOBIN targets the difficulty of achieving both powerful machine learning and very low execution latency in high-frequency trading. It reduces delay by moving inference closer to the network devices.\"},{\"question\":\"How does LOBIN generate predictions?\",\"answer\":\"LOBIN builds limit order books from high-frequency market data feeds, then performs machine-learning inference within programmable switches. It uses limit-order-book information as features for time-series prediction.\"},{\"question\":\"What performance benefits does LOBIN report compared with server-based solutions?\",\"answer\":\"LOBIN reports lower latency, higher throughput, and only minor impact on machine-learning prediction performance. It demonstrates microsecond-level latency reduction while maintaining prediction quality.\"}]","LOBIN - In-Network Machine Learning for Limit Order Books | PDF",1785672150,20,{"code":4,"msg":30,"data":31},"ok",{"site_id":23,"language":22,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"lobin-in-network-machine-learning-for-limit-order-books","",{"@graph":35,"@context":84},[36,53,67],{"@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/lobin-in-network-machine-learning-for-limit-order-books/116869/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":22,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-08-02",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What problem does LOBIN address in algorithmic trading?","Question",{"text":74,"@type":75},"LOBIN targets the difficulty of achieving both powerful machine learning and very low execution latency in high-frequency trading. It reduces delay by moving inference closer to the network devices.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does LOBIN generate predictions?",{"text":79,"@type":75},"LOBIN builds limit order books from high-frequency market data feeds, then performs machine-learning inference within programmable switches. It uses limit-order-book information as features for time-series prediction.",{"name":81,"@type":72,"acceptedAnswer":82},"What performance benefits does LOBIN report compared with server-based solutions?",{"text":83,"@type":75},"LOBIN reports lower latency, higher throughput, and only minor impact on machine-learning prediction performance. It demonstrates microsecond-level latency reduction while maintaining prediction quality.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"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":28,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":28,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]