{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T16:18:03Z","timestamp":1784132283247,"version":"3.55.0"},"reference-count":80,"publisher":"Association for Computing Machinery (ACM)","issue":"CoNEXT3","license":[{"start":{"date-parts":[[2023,11,27]],"date-time":"2023-11-27T00:00:00Z","timestamp":1701043200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100006374","name":"National Science Foundation","doi-asserted-by":"publisher","award":["CNS-2112471, CNS-1925601, and CNS-2312875"],"award-info":[{"award-number":["CNS-2112471, CNS-1925601, and CNS-2312875"]}],"id":[{"id":"10.13039\/501100006374","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100006374","name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades","doi-asserted-by":"publisher","award":["PID2021-128250NB-I00"],"award-info":[{"award-number":["PID2021-128250NB-I00"]}],"id":[{"id":"10.13039\/501100006374","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100006374","name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades","doi-asserted-by":"publisher","award":["IJC2019-039885-I"],"award-info":[{"award-number":["IJC2019-039885-I"]}],"id":[{"id":"10.13039\/501100006374","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100006374","name":"Ministerio de Universidades","doi-asserted-by":"publisher","award":["CAS21\/00500"],"award-info":[{"award-number":["CAS21\/00500"]}],"id":[{"id":"10.13039\/501100006374","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100006374","name":"Ministerio de Asuntos Econ\u00f3micos y Transformaci\u00f3n Digital, Gobierno de Espa\u00f1a","doi-asserted-by":"publisher","award":["TSI-063000-2021-63"],"award-info":[{"award-number":["TSI-063000-2021-63"]}],"id":[{"id":"10.13039\/501100006374","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["Proc. ACM Netw."],"published-print":{"date-parts":[[2023,11,27]]},"abstract":"<jats:p>The Open Radio Access Network (RAN) paradigm is transforming cellular networks into a system of disaggregated, virtualized, and software-based components. These self-optimize the network through programmable, closed-loop control, leveraging Artificial Intelligence (AI) and Machine Learning (ML) routines. In this context, Deep Reinforcement Learning (DRL) has shown great potential in addressing complex resource allocation problems. However, DRL-based solutions are inherently hard to explain, which hinders their deployment and use in practice. In this paper, we propose EXPLORA, a framework that provides explainability of DRL-based control solutions for the Open RAN ecosystem. EXPLORA synthesizes network-oriented explanations based on an attributed graph that produces a link between the actions taken by a DRL agent (i.e., the nodes of the graph) and the input state space (i.e., the attributes of each node). This novel approach allows EXPLORA to explain models by providing information on the wireless context in which the DRL agent operates. EXPLORA is also designed to be lightweight for real-time operation. We prototype EXPLORA and test it experimentally on an O-RAN-compliant near-real-time RIC deployed on the Colosseum wireless network emulator. We evaluate EXPLORA for agents trained for different purposes and showcase how it generates clear network-oriented explanations. We also show how explanations can be used to perform informative and targeted intent-based action steering and achieve median transmission bitrate improvements of 4% and tail improvements of 10%.<\/jats:p>","DOI":"10.1145\/3629141","type":"journal-article","created":{"date-parts":[[2023,11,28]],"date-time":"2023-11-28T15:40:05Z","timestamp":1701186005000},"page":"1-26","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":37,"title":["EXPLORA: AI\/ML EXPLainability for the Open RAN"],"prefix":"10.1145","volume":"1","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4323-4355","authenticated-orcid":false,"given":"Claudio","family":"Fiandrino","sequence":"first","affiliation":[{"name":"IMDEA Networks Institute, Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1511-1833","authenticated-orcid":false,"given":"Leonardo","family":"Bonati","sequence":"additional","affiliation":[{"name":"Northeastern University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7690-0449","authenticated-orcid":false,"given":"Salvatore","family":"D'Oro","sequence":"additional","affiliation":[{"name":"Northeastern University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9740-134X","authenticated-orcid":false,"given":"Michele","family":"Polese","sequence":"additional","affiliation":[{"name":"Northeastern University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2719-1789","authenticated-orcid":false,"given":"Tommaso","family":"Melodia","sequence":"additional","affiliation":[{"name":"Northeastern University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6667-8779","authenticated-orcid":false,"given":"Joerg","family":"Widmer","sequence":"additional","affiliation":[{"name":"IMDEA Networks Institute, Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2023,11,28]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11797"},{"key":"e_1_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.1145\/3300061.3345431"},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM48880.2022.9796985"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.23919\/ACC50511.2021.9483182"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2021.3129094"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/WCNC51071.2022.9771908"},{"key":"e_1_2_1_7_1","doi-asserted-by":"crossref","unstructured":"David A Broniatowski et al. 2021. Psychological foundations of explainability and interpretability in artificial intelligence. NIST Tech. Rep (2021).","DOI":"10.6028\/NIST.IR.8367"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSG.2022.3228636"},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/2939672.2939785"},{"key":"e_1_2_1_10_1","volume-title":"Explainable reinforcement learning for broad-XAI: a conceptual framework and survey. arXiv preprint arXiv:2108.09003","author":"Dazeley Richard","year":"2021","unstructured":"Richard Dazeley, Peter Vamplew, and Francisco Cruz. 2021. Explainable reinforcement learning for broad-XAI: a conceptual framework and survey. arXiv preprint arXiv:2108.09003 (2021)."},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1145\/3341216.3342210"},{"key":"e_1_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM42981.2021.9488898"},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.3390\/s21248173"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1145\/3359786"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM53939.2023.10228919"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.comcom.2022.06.036"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/MCOMSTD.101.2000014"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3570361.3613256"},{"key":"e_1_2_1_19_1","volume-title":"A survey of methods for explaining black box models. ACM computing surveys (CSUR) 51, 5","author":"Guidotti Riccardo","year":"2018","unstructured":"Riccardo Guidotti, Anna Monreale, Salvatore Ruggieri, Franco Turini, Fosca Giannotti, and Dino Pedreschi. 2018. A survey of methods for explaining black box models. ACM computing surveys (CSUR) 51, 5 (2018), 1--42."},{"key":"e_1_2_1_20_1","volume-title":"DARPA's explainable artificial intelligence (XAI) program. AI magazine 40, 2","author":"Gunning David","year":"2019","unstructured":"David Gunning and David Aha. 2019. DARPA's explainable artificial intelligence (XAI) program. AI magazine 40, 2 (2019), 44--58."},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1109\/TVT.2020.3020400"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/MCOM.001.2000050"},{"key":"e_1_2_1_23_1","volume-title":"Proc. of EuroMLSys.","author":"Hamadanian Pouya","year":"2022","unstructured":"Pouya Hamadanian, Malte Schwarzkopf, Siddartha Sen, and Mohammad Alizadeh. 2022. How Reinforcement Learning Systems Fail and What to do About It. In Proc. of EuroMLSys."},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1145\/3460120.3484589"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ast.2021.107052"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/MCI.2021.3129959"},{"key":"e_1_2_1_27_1","first-page":"7","article-title":"Joint Server Selection, Cooperative Offloading and Handover in Multi-Access Edge Computing Wireless Network: A Deep Reinforcement Learning Approach","volume":"21","author":"Ho Tai Manh","year":"2022","unstructured":"Tai Manh Ho and Kim-Khoa Nguyen. 2022. Joint Server Selection, Cooperative Offloading and Handover in Multi-Access Edge Computing Wireless Network: A Deep Reinforcement Learning Approach. IEEE Transactions on Mobile Computing 21, 7 (July 2022), 2421--2435.","journal-title":"IEEE Transactions on Mobile Computing"},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2023.3243068"},{"key":"e_1_2_1_29_1","unstructured":"Intel. 2022. Intelligent Connection Management for Automated Handover Reference Implementation. https:\/\/www.intel.com\/content\/www\/us\/en\/developer\/articles\/reference-implementation\/intelligent-connection-management.html."},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.3390\/s22145375"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1145\/3548606.3560609"},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.1145\/3477086.3480842"},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1145\/3341216.3342218"},{"key":"e_1_2_1_34_1","volume-title":"Leveraging Applications of Formal Methods, Verification and Validation: Verification Principles","author":"K\u00f6nighofer Bettina","unstructured":"Bettina K\u00f6nighofer, Florian Lorber, Nils Jansen, and Roderick Bloem. 2020. Shield Synthesis for Reinforcement Learning. In Leveraging Applications of Formal Methods, Verification and Validation: Verification Principles, Tiziana Margaria and Bernhard Steffen (Eds.). Springer International Publishing, 290--306."},{"key":"e_1_2_1_35_1","unstructured":"M. Korobov and K. Lopuhin. 2021. ELI5 is a Python library - v. 0.11. Available at (accessed 26\/10\/2021): https:\/\/eli5.readthedocs.io\/en\/latest\/."},{"key":"e_1_2_1_36_1","volume-title":"Explainability in reinforcement learning: perspective and position. arXiv preprint arXiv:2203.11547","author":"Krajna Agneza","year":"2022","unstructured":"Agneza Krajna, Mario Brcic, Tomislav Lipic, and Juraj Doncevic. 2022. Explainability in reinforcement learning: perspective and position. arXiv preprint arXiv:2203.11547 (2022)."},{"key":"e_1_2_1_37_1","volume-title":"United States. Department of Energy. Office of Scientific, and Technical Information","author":"Lawrence Berkeley National Laboratory","year":"2013","unstructured":"Lawrence Berkeley National Laboratory, United States. Department of Energy. Office of Scientific, and Technical Information. 2013. Attributed Graph Models: Modeling Network Structure with Correlated Attributes. United States. Department of Energy."},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN.2010.5596468"},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2022.3215919"},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1109\/MVT.2020.3017181"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1145\/3447548.3467420"},{"key":"e_1_2_1_42_1","unstructured":"Peizheng Li Jonathan Thomas Xiaoyang Wang Ahmed Khalil Abdelrahim Ahmad Rui Inacio Shipra Kapoor Arjun Parekh Angela Doufexi Arman Shojaeifard et al. 2021. RLOps: Development Life-cycle of Reinforcement Learning Aided Open RAN. arXiv preprint (2021)."},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM53939.2023.10229031"},{"key":"e_1_2_1_44_1","doi-asserted-by":"publisher","DOI":"10.1145\/3341302.3342221"},{"key":"e_1_2_1_45_1","volume-title":"From local explanations to global understanding with explainable AI for trees. Nature machine intelligence 2, 1","author":"Lundberg Scott M","year":"2020","unstructured":"Scott M Lundberg, Gabriel Erion, Hugh Chen, Alex DeGrave, Jordan M Prutkin, Bala Nair, Ronit Katz, Jonathan Himmelfarb, Nisha Bansal, and Su-In Lee. 2020. From local explanations to global understanding with explainable AI for trees. Nature machine intelligence 2, 1 (2020), 56--67."},{"key":"e_1_2_1_46_1","volume-title":"Proc. of NIPS. 4768--4777","author":"Lundberg Scott M","year":"2017","unstructured":"Scott M Lundberg and Su-In Lee. 2017. A unified approach to interpreting model predictions. In Proc. of NIPS. 4768--4777."},{"key":"e_1_2_1_47_1","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2019.2916583"},{"key":"e_1_2_1_48_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i03.5631"},{"key":"e_1_2_1_49_1","doi-asserted-by":"publisher","DOI":"10.1145\/3452296.3472906"},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.1145\/3098822.3098843"},{"key":"e_1_2_1_51_1","unstructured":"Mavenir. 2023. Building the World's First O-RAN-Compliant AI-Powered Closed-Loop Near-RT RIC. https:\/\/www.mavenir.com\/blog\/building-the-worlds-first-o-ran-compliant-ai-powered-closed-loop-near-rt-ric\/."},{"key":"e_1_2_1_52_1","doi-asserted-by":"publisher","DOI":"10.1145\/3447993.3488032"},{"key":"e_1_2_1_53_1","doi-asserted-by":"publisher","DOI":"10.1145\/3387514.3405859"},{"key":"e_1_2_1_54_1","doi-asserted-by":"publisher","DOI":"10.1145\/3511597"},{"key":"e_1_2_1_55_1","doi-asserted-by":"publisher","DOI":"10.1007\/978--3-030--28954--6_10"},{"key":"e_1_2_1_56_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.comcom.2022.05.013"},{"key":"e_1_2_1_57_1","volume-title":"Proc. of USENIX NSDI. 1135--1158","author":"Naseer Usama","year":"2022","unstructured":"Usama Naseer and Theophilus A Benson. 2022. Configanator: A Data-driven Approach to Improving CDN Performance. In Proc. of USENIX NSDI. 1135--1158."},{"key":"e_1_2_1_58_1","volume-title":"d.]. Amber Release. https:\/\/wiki.o-ran-sc.org\/pages\/viewpage.action?pageId=14221337. Accessed","author":"Software Community O-RAN","year":"2020","unstructured":"O-RAN Software Community. [n. d.]. Amber Release. https:\/\/wiki.o-ran-sc.org\/pages\/viewpage.action?pageId=14221337. Accessed July 2020."},{"key":"e_1_2_1_59_1","unstructured":"O-RAN Working Group 3. 2021. O-RAN Near-RT RAN Intelligent Controller Near-RT RIC Architecture 2.00. O-RAN.WG3.RICARCH-v02.00."},{"key":"e_1_2_1_60_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2022.3188013"},{"key":"e_1_2_1_61_1","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2023.3239220"},{"key":"e_1_2_1_62_1","doi-asserted-by":"publisher","DOI":"10.1145\/3299874.3319493"},{"key":"e_1_2_1_63_1","volume-title":"Veith","author":"Puiutta Erika","year":"2020","unstructured":"Erika Puiutta and Eric M. S. P. Veith. 2020. Explainable Reinforcement Learning: A Survey. In Machine Learning and Knowledge Extraction, Andreas Holzinger, Peter Kieseberg, A Min Tjoa, and EdgarWeippl (Eds.). Springer International Publishing, 77--95."},{"key":"e_1_2_1_64_1","doi-asserted-by":"publisher","DOI":"10.1145\/2939672.2939778"},{"key":"e_1_2_1_65_1","doi-asserted-by":"publisher","DOI":"10.1109\/CCGrid54584.2022.00113"},{"key":"e_1_2_1_66_1","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM53939.2023.10228989"},{"key":"e_1_2_1_67_1","volume-title":"Proc. of ICML. 3145--3153","author":"Shrikumar Avanti","year":"2017","unstructured":"Avanti Shrikumar, Peyton Greenside, and Anshul Kundaje. 2017. Learning important features through propagating activation differences. In Proc. of ICML. 3145--3153."},{"key":"e_1_2_1_68_1","doi-asserted-by":"publisher","DOI":"10.1109\/MILCOM52596.2021.9653140"},{"key":"e_1_2_1_69_1","volume-title":"Reinforcement learning: An introduction","author":"Sutton Richard S","year":"2018","unstructured":"Richard S Sutton and Andrew G Barto. 2018. Reinforcement learning: An introduction second edition. Adaptive computation and machine learning: The MIT Press, Cambridge, MA and London (2018)."},{"key":"e_1_2_1_70_1","unstructured":"Unwired Labs. Accessed March 2023. OpenCelliD. https:\/\/opencellid.org."},{"key":"e_1_2_1_71_1","volume-title":"d.]. MGEN Traffic Emulator. https:\/\/www.nrl.navy.mil\/Our-Work\/Areas-of- Research\/Information-Technology\/NCS\/MGEN. Accessed","author":"Naval Research Laboratory U.S.","year":"2021","unstructured":"U.S. Naval Research Laboratory. [n. d.]. MGEN Traffic Emulator. https:\/\/www.nrl.navy.mil\/Our-Work\/Areas-of- Research\/Information-Technology\/NCS\/MGEN. Accessed September 2021."},{"key":"e_1_2_1_72_1","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.2112.04698"},{"key":"e_1_2_1_73_1","doi-asserted-by":"publisher","DOI":"10.1109\/ISORC52572.2022.9812837"},{"key":"e_1_2_1_74_1","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2018.8485853"},{"key":"e_1_2_1_75_1","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2022.3222314"},{"key":"e_1_2_1_76_1","doi-asserted-by":"publisher","DOI":"10.1109\/MCOM.2017.1700400"},{"key":"e_1_2_1_77_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCSS.2021.3096824"},{"key":"e_1_2_1_78_1","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM48880.2022.9796791"},{"key":"e_1_2_1_79_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSAC.2021.3126085"},{"key":"e_1_2_1_80_1","doi-asserted-by":"publisher","DOI":"10.1145\/3232565.3232569"}],"container-title":["Proceedings of the ACM on Networking"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3629141","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3629141","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,20]],"date-time":"2026-04-20T18:46:15Z","timestamp":1776710775000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3629141"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,11,27]]},"references-count":80,"journal-issue":{"issue":"CoNEXT3","published-print":{"date-parts":[[2023,11,27]]}},"alternative-id":["10.1145\/3629141"],"URL":"https:\/\/doi.org\/10.1145\/3629141","relation":{},"ISSN":["2834-5509"],"issn-type":[{"value":"2834-5509","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,11,27]]},"assertion":[{"value":"2023-11-28","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}