{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T04:59:01Z","timestamp":1750309141714,"version":"3.41.0"},"reference-count":52,"publisher":"Association for Computing Machinery (ACM)","issue":"2","license":[{"start":{"date-parts":[[2024,5,13]],"date-time":"2024-05-13T00:00:00Z","timestamp":1715558400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM J. Comput. Sustain. Soc."],"published-print":{"date-parts":[[2024,6,30]]},"abstract":"<jats:p>\n            Developing countries need to better manage fast increasing traffic flows, owing to rapid urbanization. Else, increasing traffic congestion would increase fatalities due to reckless driving, as well as keep vehicular emissions and air pollution critically high in cities like New Delhi. State-of-the-art traffic signal control methods in developed countries, however, use expensive sensing, computation, and communication resources. How far can control algorithms go, under resource constraints, is explored through the design and evaluation of\n            <jats:sc>FrugalLight<\/jats:sc>\n            (FL) in this article. We also captured and processed a real traffic dataset at a busy intersection in New Delhi, India, using efficient techniques on low cost embedded devices. This dataset (\n            <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" ext-link-type=\"uri\" xlink:href=\"https:\/\/delhi-trafficdensity-dataset.github.io\">https:\/\/delhi-trafficdensity-dataset.github.io<\/jats:ext-link>\n            ) contains traffic density information at fine time granularity of one measurement every second, from all approaches of the intersection for 40 days.\n            <jats:sc>FrugalLight<\/jats:sc>\n            (\n            <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" ext-link-type=\"uri\" xlink:href=\"https:\/\/github.com\/sachin-iitd\/FrugalLight\">https:\/\/github.com\/sachin-iitd\/FrugalLight<\/jats:ext-link>\n            ) is evaluated on the collected traffic dataset from New Delhi and another open source traffic dataset from New York.\n            <jats:sc>FrugalLight<\/jats:sc>\n            matches the performance of state-of-the-art Convolutional Neural Network (CNN) based sensing and Deep Reinforcement Learning (DRL) based control algorithms, while utilizing resources less by an order of magnitude. We further explore improvements using a careful combination of knowledge distillation and domain knowledge based DRL model compression, with employing Model-Agnostic Meta-Learning to quickly adapt to traffic at new intersections. The collected real dataset and\n            <jats:sc>FrugalLight<\/jats:sc>\n            therefore opens up opportunities for resource efficient RL based intersection control design for the ML research community, where the controller should have limited carbon footprint. Such intelligent, green, intersection controllers can help reduce traffic congestion and associated vehicular emissions, even if compute and communication infrastructure is limited in low resource regions. This is a critical step toward achieving two of the United Nations Sustainable Development Goals (SDG), namely sustainable cities and communities and climate action.\n          <\/jats:p>","DOI":"10.1145\/3648599","type":"journal-article","created":{"date-parts":[[2024,2,19]],"date-time":"2024-02-19T12:20:42Z","timestamp":1708345242000},"page":"1-32","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["F\n            <scp>rugal<\/scp>\n            L\n            <scp>ight<\/scp>\n            : Symmetry-Aware Cyclic Heterogeneous Intersection Control using Deep Reinforcement Learning with Model Compression, Distillation and Domain Knowledge"],"prefix":"10.1145","volume":"2","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6773-0989","authenticated-orcid":false,"given":"Sachin Kumar","family":"Chauhan","sequence":"first","affiliation":[{"name":"CSE, Indian Institute of Technology Delhi, New Delhi, India"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2465-3650","authenticated-orcid":false,"given":"Rijurekha","family":"Sen","sequence":"additional","affiliation":[{"name":"CSE, Indian Institute of Technology Delhi, New Delhi, India"}]}],"member":"320","published-online":{"date-parts":[[2024,5,13]]},"reference":[{"key":"e_1_3_2_2_2","unstructured":"2021. Meta learning (computer science). (2021). Retrieved from https:\/\/en.wikipedia.org\/wiki\/Meta_learning_(computer_science)"},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.trc.2017.09.020"},{"key":"e_1_3_2_4_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.aei.2018.08.002"},{"key":"e_1_3_2_5_2","unstructured":"U. K. Bewiser. 2016. Traffic lights cause traffic jams new research suggests. (2016). Retrieved from https:\/\/www.bewiser.co.uk\/news\/car-insurance\/traffic-lights-cause-traffic-jams-new-research-suggests"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.deveng.2022.100105"},{"key":"e_1_3_2_7_2","unstructured":"Thierry Bouwmans. 2014. Background modeling and Foreground Detection for video surveillance: Traditional and Recent Approaches Benchmarking and Evaluation. (2014). Retrieved from http:\/\/www.crcpress.com\/product\/isbn\/9781482205374."},{"key":"e_1_3_2_8_2","unstructured":"CC-by4. 2013. Attribution 4.0 International (CC BY 4.0). (2013). Retrieved June 7 2023 from https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"key":"e_1_3_2_9_2","doi-asserted-by":"publisher","DOI":"10.1145\/3287098.3287118"},{"key":"e_1_3_2_10_2","article-title":"EcoLight: Intersection control in developing regions under extreme budget and network constraints","author":"Chauhan Sachin","year":"2020","unstructured":"Sachin Chauhan, Kashish Bansal, and Rijurekha Sen. 2020. EcoLight: Intersection control in developing regions under extreme budget and network constraints. In Proceedings of the 34th Conference on Neural Information Processing Systems. Retrieved from https:\/\/proceedings.neurips.cc\/paper\/2020\/file\/97e49161287e7a4f9b745366e4f9431b-Paper.pdf","journal-title":"Proceedings of the 34th Conference on Neural Information Processing Systems"},{"key":"e_1_3_2_11_2","article-title":"AirDelhi: Fine-grained spatio-temporal particulate matter dataset from delhi for ML based modeling","author":"Chauhan Sachin","year":"2023","unstructured":"Sachin Chauhan, Sayan Ranu, Rijurekha Sen, Zeel B. Patel, and Nipun Batra. 2023. AirDelhi: Fine-grained spatio-temporal particulate matter dataset from delhi for ML based modeling. In Proceedings of the 37th Conference on Neural Information Processing Systems Track on Datasets and Benchmarks. Retrieved from https:\/\/openreview.net\/pdf?id=n2wW7goGky","journal-title":"Proceedings of the 37th Conference on Neural Information Processing Systems Track on Datasets and Benchmarks"},{"key":"e_1_3_2_12_2","volume-title":"Proceedings of the NeurIPS 2023 Computational Sustainability: Promises and Pitfalls from Theory to Deployment","author":"Chauhan Sachin","year":"2023","unstructured":"Sachin Chauhan and Rijurekha Sen. 2023. RealLight: DRL based intersection control in developing countries without traffic simulators. In Proceedings of the NeurIPS 2023 Computational Sustainability: Promises and Pitfalls from Theory to Deployment. Retrieved from https:\/\/openreview.net\/pdf?id=dmjT841VuV"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-84628-982-8_3"},{"key":"e_1_3_2_14_2","unstructured":"Anthony Davis. 2019. Inductive loops or wireless magnetometers for traffic signal control. (2019). Retrieved from https:\/\/highways.today\/2019\/03\/22\/inductive-loops-wireless-magnetometers"},{"key":"e_1_3_2_15_2","doi-asserted-by":"publisher","DOI":"10.1109\/ITSC.2010.5625066"},{"key":"e_1_3_2_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/TITS.2013.2255286"},{"key":"e_1_3_2_17_2","unstructured":"FHWA. 2006. Traffic Detector Handbook: Chapter 4. In-Roadway Sensor Design. (2006). Retrieved from https:\/\/www.fhwa.dot.gov\/publications\/research\/operations\/its\/06108\/04.cfm)"},{"key":"e_1_3_2_18_2","unstructured":"Chelsea Finn Pieter Abbeel and Sergey Levine. 2017. Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks. (2017). arXiv:1703.03400. Retrieved from https:\/\/arxiv.org\/abs\/1703.03400"},{"key":"e_1_3_2_19_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01103"},{"key":"e_1_3_2_20_2","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2018.2808319"},{"key":"e_1_3_2_21_2","doi-asserted-by":"publisher","DOI":"10.1080\/03081060903429421"},{"key":"e_1_3_2_22_2","unstructured":"Song Han Jeff Pool John Tran and William J. Dally. 2015. Learning both Weights and Connections for Efficient Neural Networks. (2015). arXiv:1506.02626. Retrieved from https:\/\/arxiv.org\/abs\/1506.02626"},{"issue":"8","key":"e_1_3_2_23_2","first-page":"2","article-title":"Neural networks for machine learning lecture 6a overview of mini-batch gradient descent","volume":"14","author":"Hinton Geoffrey","year":"2012","unstructured":"Geoffrey Hinton, Nitish Srivastava, and Kevin Swersky. 2012. Neural networks for machine learning lecture 6a overview of mini-batch gradient descent. Cited on 14, 8 (2012), 2.","journal-title":"Cited on"},{"key":"e_1_3_2_24_2","first-page":"3772","volume-title":"Proceedings of the ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing.","author":"Iyer Shiva R.","year":"2020","unstructured":"Shiva R. Iyer, Ulzee An, and Lakshminarayanan Subramanian. 2020. Forecasting sparse traffic congestion patterns using message-passing rnns. In Proceedings of the ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing.IEEE, 3772\u20133776."},{"key":"e_1_3_2_25_2","doi-asserted-by":"publisher","DOI":"10.1145\/2160601.2160616"},{"key":"e_1_3_2_26_2","unstructured":"Nicole Kobie. 2018. London is hacking its traffic lights to slash waiting times. (2018). Retrieved from https:\/\/www.wired.co.uk\/article\/traffic-lights-uk-london"},{"key":"e_1_3_2_27_2","first-page":"183","article-title":"SUMO (simulation of urban mobility); An open-source traffic simulation","author":"Krajzewicz Daniel","year":"2002","unstructured":"Daniel Krajzewicz, Georg Hertkorn, Christian Feld, and Peter Wagner. 2002. SUMO (simulation of urban mobility); An open-source traffic simulation. In Proceedings of the 4th Middle East Symposium on Simulation and Modelling. 183\u2013187.","journal-title":"Proceedings of the 4th Middle East Symposium on Simulation and Modelling"},{"key":"e_1_3_2_28_2","first-page":"7517","volume-title":"Proceedings of the Advances in Neural Information Processing Systems 31.","author":"lan Xu","year":"2018","unstructured":"Xu lan, Xiatian Zhu, and Shaogang Gong. 2018. Knowledge distillation by on-the-fly native ensemble. In Proceedings of the Advances in Neural Information Processing Systems 31.S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett (Eds.), Curran Associates, Inc., 7517\u20137527. Retrieved from http:\/\/papers.nips.cc\/paper\/7980-knowledge-distillation-by-on-the-fly-native-ensemble.pdf"},{"key":"e_1_3_2_29_2","unstructured":"John D. C. Little M. D. Kelson and N. H. Gartner. 1981. MAXBAND: A versatile program for setting signals on arteries and triangular networks. (1981)."},{"key":"e_1_3_2_30_2","unstructured":"P. Lowrie. 1990. SCATS: Sydney co-ordinated adaptive traffic system: A traffic responsive method of controlling urban traffic."},{"key":"e_1_3_2_31_2","doi-asserted-by":"crossref","unstructured":"Huizi Mao Song Han Jeff Pool Wenshuo Li Xingyu Liu Yu Wang and William J. Dally. 2017. Exploring the Regularity of Sparse Structure in Convolutional Neural Networks. (2017). arXiv:1705.08922. Retrieved from https:\/\/arxiv.org\/abs\/1705.08922","DOI":"10.1109\/CVPRW.2017.241"},{"key":"e_1_3_2_32_2","article-title":"On student-teacher deviations in distillation: Does it pay to disobey","author":"Nagarajan Vaishnavh","year":"2023","unstructured":"Vaishnavh Nagarajan, Aditya Krishna Menon, Srinadh Bhojanapalli, Hossein Mobahi, and Sanjiv Kumar. 2023. On student-teacher deviations in distillation: Does it pay to disobey. In Proceedings of the 37th Conference on Neural Information Processing Systems.","journal-title":"Proceedings of the 37th Conference on Neural Information Processing Systems"},{"key":"e_1_3_2_33_2","doi-asserted-by":"publisher","DOI":"10.1109\/ITSC.2018.8569301"},{"key":"e_1_3_2_34_2","unstructured":"OpenCV. 2021. Optical Flow. (2021). Retrieved Oct 29 2021 fromhttps:\/\/docs.opencv.org\/3.4.15\/d4\/dee\/tutorial_optical_flow.html"},{"key":"e_1_3_2_35_2","article-title":"Counting device selection and reliability: Synthesis study","author":"Peeta S.","year":"2002","unstructured":"S. Peeta and P. Zhang. 2002. Counting device selection and reliability: Synthesis study. Joint Transportation Research Program (2002).","journal-title":"Joint Transportation Research Program"},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/ITSC.2011.6082823"},{"key":"e_1_3_2_37_2","doi-asserted-by":"publisher","unstructured":"Joseph Redmon Santosh Divvala Ross Girshick and Ali Farhadi. 2016. You only look once: Unified real-time object detection. 779\u2013788. DOI:10.1109\/CVPR.2016.91","DOI":"10.1109\/CVPR.2016.91"},{"key":"e_1_3_2_38_2","doi-asserted-by":"publisher","DOI":"10.1145\/3292500.3330988"},{"key":"e_1_3_2_39_2","volume-title":"Traffic Engineering","author":"Roess Roger P.","year":"2004","unstructured":"Roger P. Roess, Elena S. Prassas, and William R. McShane. 2004. Traffic Engineering. Pearson\/Prentice Hall. 04295239Retrieved from https:\/\/books.google.co.in\/books?id=OYNPAAAAMAAJ"},{"key":"e_1_3_2_40_2","doi-asserted-by":"publisher","DOI":"10.23919\/ACC55779.2023.10156258"},{"key":"e_1_3_2_41_2","doi-asserted-by":"publisher","DOI":"10.1145\/3530190.3534823"},{"key":"e_1_3_2_42_2","unstructured":"Elise van der Pol and Frans A. Oliehoek. 2016. Coordinated deep reinforcement learners for traffic light control. Retrieved from https:\/\/api.semanticscholar.org\/CorpusID:198950131"},{"key":"e_1_3_2_43_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.trc.2013.08.014"},{"key":"e_1_3_2_44_2","unstructured":"Shiva Verma. 2019. Understanding different Loss Functions for Neural Networks. (2019). Retrieved from https:\/\/towardsdatascience.com\/loss-functions-neural-networks-dd1ed0274718"},{"key":"e_1_3_2_45_2","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2020.3033782"},{"key":"e_1_3_2_46_2","doi-asserted-by":"publisher","DOI":"10.1145\/3292500.3330949"},{"key":"e_1_3_2_47_2","doi-asserted-by":"publisher","DOI":"10.1145\/3357384.3357902"},{"key":"e_1_3_2_48_2","unstructured":"Hua Wei Guanjie Zheng Vikash Gayah and Zhenhui Li. 2019. A Survey on Traffic Signal Control Methods. (2019). arXiv:1904.08117. Retrieved from https:\/\/arxiv.org\/abs\/1904.08117"},{"key":"e_1_3_2_49_2","unstructured":"David Williams. 2012. Too many traffic lights make congestion worse. (2012). Retrieved from https:\/\/www.standard.co.uk\/hp\/front\/too-many-traffic-lights-make-congestion-worse-6676646.html)"},{"key":"e_1_3_2_50_2","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr.2017.643"},{"key":"e_1_3_2_51_2","doi-asserted-by":"publisher","DOI":"10.1145\/3140659.3080215"},{"key":"e_1_3_2_52_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i01.5467"},{"key":"e_1_3_2_53_2","doi-asserted-by":"publisher","DOI":"10.1145\/3308558.3314139"}],"container-title":["ACM Journal on Computing and Sustainable Societies"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3648599","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3648599","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T22:50:20Z","timestamp":1750287020000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3648599"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,5,13]]},"references-count":52,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2024,6,30]]}},"alternative-id":["10.1145\/3648599"],"URL":"https:\/\/doi.org\/10.1145\/3648599","relation":{},"ISSN":["2834-5533"],"issn-type":[{"type":"electronic","value":"2834-5533"}],"subject":[],"published":{"date-parts":[[2024,5,13]]},"assertion":[{"value":"2024-01-09","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-01-26","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-05-13","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}