Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Review Article
  • Published:

Artificial intelligence for low-carbon energy and information networks

Nature Reviews Electrical Engineering volume 3, pages 238–253 (2026) Cite this article

Subjects

Abstract

Achieving low-carbon energy and information networks requires coordination between variable renewable energy supply and fluctuating traffic demand. Artificial intelligence (AI) systems offer tools both to optimize networks and to coordinate between them. Yet AI systems generate substantial carbon emissions — from model training, deployment and use, and the hardware life cycle — creating a paradox in which the solution contributes to the problem. This Review analyses the dual role of AI systems. We examine AI applications in energy supply networks and information networks, the capabilities required for supply–demand coordination, and the carbon emissions of AI systems themselves. Achieving low-carbon AI systems has become a central challenge to ensure that their environmental benefits outweigh their costs. Looking forward, we outline three research directions: low-carbon AI goals, energy-efficient large models, and AI-driven life-cycle management. Furthermore, we call for policy frameworks and industry strategies to achieve low-carbon energy and information networks.

Key points

  • Artificial intelligence (AI) plays a dual role in network decarbonization: optimizing energy and information networks while generating its own carbon emissions.

  • In energy supply networks, AI forecasts renewable generation and optimizes grid integration. In information networks, AI enables energy-efficient routing, resource allocation and predictive maintenance.

  • Coordinating variable energy supply with fluctuating traffic demand requires four AI capabilities: spatiotemporal prediction, real-time decision-making, multi-objective optimization, and scalable solutions.

  • Carbon emissions of AI arise from training, deployment and the hardware life cycle; generative AI amplifies these through energy-intensive training and continuous inference.

  • Achieving low-carbon AI requires advances across network operations, large model efficiency and equipment life-cycle management, supported by policy frameworks and industry strategies.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to the full article PDF.

USD 39.95

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: AI systems and energy trends.
The alternative text for this image may have been generated using AI.
Fig. 2: AI systems for low-carbon energy supply networks.
The alternative text for this image may have been generated using AI.
Fig. 3: AI methods for coordinating energy supply and traffic demand.
The alternative text for this image may have been generated using AI.
Fig. 4: Reducing carbon emissions in AI training.
The alternative text for this image may have been generated using AI.
Fig. 5: Reducing carbon emissions during AI deployment and use.
The alternative text for this image may have been generated using AI.
Fig. 6: Reducing hardware-related carbon emissions of AI methods.
The alternative text for this image may have been generated using AI.

Similar content being viewed by others

References

  1. Saud, S., Haseeb, A., Chen, S. & Li, H. The role of information and communication technology and financial development in shaping a low-carbon environment: a belt and road journey toward development. Inform. Technol. Dev. 29, 83–102 (2023).

    Article  Google Scholar 

  2. Masciari, E. & Napolitano, E. V. The environmental cost of high performance computing system simulation. In 2024 32nd Euromicro Int. Conf. Parallel, Distributed and Network-Based Processing (PDP) (eds Chis, A. E. et al.) 289–292 (IEEE, 2024).

  3. Li, T. et al. Carbon emissions of 5G mobile networks in China. Nat. Sustain. 6, 1620–1631 (2023). This study develops a data-driven framework to characterize a carbon efficiency trap within China’s 5G networks. The authors propose DeepEnergy, an energy-saving method, to reduce carbon emissions and help the network achieve more than 50% of its net-zero goal by 2023.

    Article  Google Scholar 

  4. Shobande, O. A. & Asongu, S. A. Searching for sustainable footprints: does ICT increase CO₂ emissions? Environ. Model. Assess. 28, 133–143 (2023).

    Article  Google Scholar 

  5. Israr, A., Yang, Q., Li, W. & Zomaya, A. Y. Renewable energy powered sustainable 5G network infrastructure: opportunities, challenges and perspectives. J. Netw. Comput. Appl. 175, 102910 (2021).

    Article  Google Scholar 

  6. Auer, G. et al. How much energy is needed to run a wireless network? IEEE Wirel. Commun. 18, 40–49 (2011).

    Article  Google Scholar 

  7. Kabeyi, M. J. B. & Olanrewaju, O. A. Sustainable energy transition for renewable and low carbon grid electricity generation and supply. Front. Energy Res. 9, 743114 (2022).

    Article  Google Scholar 

  8. International Renewable Energy Agency (IRENA). Renewable Capacity Statistics 2025 (IRENA, 2025).

  9. Pan, S. L. & Nishant, R. Artificial intelligence for digital sustainability: an insight into domain-specific research and future directions. Int. J. Inf. Manage. 72, 102668 (2023).

    Google Scholar 

  10. Ahmad, T. et al. Artificial intelligence in sustainable energy industry: status quo, challenges and opportunities. J. Clean. Prod. 289, 125834 (2021).

    Article  Google Scholar 

  11. Mao, B. et al. AI models for green communications towards 6G. IEEE Commun. Surv. Tutor. 24, 210–247 (2022). This paper surveys AI-based approaches for green communications, highlighting how AI, particularly machine learning, can manage networks, improve energy efficiency and address the increasing energy demands of 5G and 6G, while discussing challenges and open research issues for sustainable 6G development.

    Article  Google Scholar 

  12. Li, R. et al. Intelligent 5G: when cellular networks meet artificial intelligence. IEEE Wirel. Commun. 24, 175–183 (2017). This article explores the emergence of native intelligence in key aspects of 5G cellular networks and emphasizes the need for full AI integration to manage complex configurations and new service demands, highlighting opportunities and challenges for AI to orchestrate intelligent 5G networks and realize information and communication technology as a primary enabler.

    Article  Google Scholar 

  13. Ge, L. & Li, Y. in Smart Power Distribution Network: Situation Awareness, Planning, and Operation (eds Ge, L. & Li, Y.) 3–17 (Springer Nature, 2023). This book chapter provides a comprehensive overview of smart power distribution networks, focusing on situational awareness, planning and operational strategies to enhance efficiency, reliability and adaptability in modern power systems.

  14. Walia, G. K., Kumar, M. & Gill, S. S. AI-empowered fog/edge resource management for IoT applications: a comprehensive review, research challenges and future perspectives. IEEE Commun. Surv. Tutor. 26, 619–669 (2023).

    Article  Google Scholar 

  15. Hua, H. et al. Edge computing with artificial intelligence: a machine learning perspective. ACM Comput. Surv. 55, 1–35 (2023).

    Article  Google Scholar 

  16. Meng, Y. et al. Transmission and distribution network-constrained large-scale demand response based on locational customer directrix load for accommodating renewable energy. Appl. Energy 350, 121681 (2023).

    Article  Google Scholar 

  17. Zhao, Q., Li, G., Cai, J., Zhou, M. & Feng, L. A tutorial on internet of behaviors: concept, architecture, technology, applications, and challenges. IEEE Commun. Surv. Tutor. 25, 1227–1260 (2023).

    Article  Google Scholar 

  18. Notton, G. et al. Intermittent and stochastic character of renewable energy sources: consequences, cost of intermittence and benefit of forecasting. Renew. Sustain. Energy Rev. 87, 96–105 (2018).

    Article  Google Scholar 

  19. Purkait, P., Basu, M. & Nath, S. R. in Challenges and Opportunities of Distributed Renewable Power (eds De, S. et al.) 37–100 (Springer, 2024). This book chapter explores the integration of renewable energy sources into existing power grids, discussing the opportunities, challenges and solutions for effective integration while highlighting recent advances in simulation tools, mathematical models and technologies such as power electronics, information and communication technology, and smart grids to support a sustainable energy future.

  20. Hamdan, A., Ibekwe, K. I., Ilojianya, V. I., Sonko, S. & Etukudoh, E. A. AI in renewable energy: a review of predictive maintenance and energy optimization. Int. J. Sci. Res. Arch. 11, 718–729 (2024).

    Article  Google Scholar 

  21. Chen, L., Li, X. & Zhu, J. Carbon peak control for achieving net-zero renewable-based smart cities: digital twin modeling and simulation. Sustain. Energy Technol. Assess. 65, 103792 (2024).

    Google Scholar 

  22. Gaur, L., Afaq, A., Arora, G. K. & Khan, N. Artificial intelligence for carbon emissions using system of systems theory. Ecol. Inform. 76, 102165 (2023). This study explores the dual impact of AI on the environment, analysing the complex relationship between the potential of AI to combat climate change and its contribution to carbon emissions, and advocates for sustainable AI practices throughout its life cycle to balance efficiency with environmental responsibility.

    Article  Google Scholar 

  23. Cowls, J., Tsamados, A., Taddeo, M. & Floridi, L. The AI gambit: leveraging artificial intelligence to combat climate change—opportunities, challenges, and recommendations. AI Soc. 38, 283–307 (2023).

    Article  Google Scholar 

  24. Alzoubi, Y. I. & Mishra, A. Green artificial intelligence initiatives: potentials and challenges. J. Clean Prod. 468, 143090 (2024).

    Article  Google Scholar 

  25. Wang, H., Lei, Z., Zhang, X., Zhou, B. & Peng, J. A review of deep learning for renewable energy forecasting. Energy Conv. Manag. 198, 111799 (2019).

    Article  Google Scholar 

  26. Yang, T., Zhao, L., Li, W. & Zomaya, A. Y. Dynamic energy dispatch strategy for integrated energy system based on improved deep reinforcement learning. Energy 235, 121377 (2021).

    Article  Google Scholar 

  27. Zhu, Z., Sun, C., He, Y., Shen, J. & Sun, J. Layout methods for integrated energy supply service stations from the perspective of combination optimization. J. Adv. Transp. 2021, 6664115 (2021).

    Article  Google Scholar 

  28. Sinsel, S. R., Riemke, R. L. & Hoffmann, V. H. Challenges and solution technologies for the integration of variable renewable energy sources—a review. Renew. Energy 145, 2271–2285 (2020).

    Article  Google Scholar 

  29. Adefarati, T. & Bansal, R. C. Integration of renewable distributed generators into the distribution system: a review. IET Renew. Power Gener. 10, 873–884 (2016).

    Article  Google Scholar 

  30. Yang, T., Zhao, L., Li, W. & Zomaya, A. Y. Reinforcement learning in sustainable energy and electric systems: a survey. Annu. Rev. Control. 49, 145–163 (2020).

    Article  MathSciNet  Google Scholar 

  31. Arumugham, V. et al. An artificial-intelligence-based renewable energy prediction program for demand-side management in smart grids. Sustainability 15, 5453 (2023). This paper presents a model for accurately predicting and managing renewable energy supply in smart grids, using prediction models and demand response programmes to optimize energy scheduling and reduce operational costs, with results validated through a multi-objective ant colony optimization algorithm.

    Article  Google Scholar 

  32. Boza, P. & Evgeniou, T. Artificial intelligence to support the integration of variable renewable energy sources to the power system. Appl. Energy 290, 116754 (2021).

    Article  Google Scholar 

  33. Nemitallah, M. A. et al. Artificial intelligence for control and optimization of boilers’ performance and emissions: a review. J. Clean. Prod. 417, 138109 (2023).

    Article  Google Scholar 

  34. Aliramezani, M., Koch, C. R. & Shahbakhti, M. Modeling, diagnostics, optimization, and control of internal combustion engines via modern machine learning techniques: a review and future directions. Prog. Energy Combust. Sci. 88, 100967 (2022).

    Article  Google Scholar 

  35. Lee, J. et al. Intelligent maintenance systems and predictive manufacturing. J. Manuf. Sci. Eng. 142, 110805–110827 (2020).

    Article  Google Scholar 

  36. Wong, K. P. & Cheung, H. N. Thermal generator scheduling algorithm based on heuristic-guided depth-first search. IEE Proc. C. Gener. Transm. Distrib. UK 137, 33 (1990).

    Article  Google Scholar 

  37. Cao, D. et al. Reinforcement learning and its applications in modern power and energy systems: a review. J. Mod. Power Syst. Clean. Energy. 8, 1029–1042 (2020).

    Article  Google Scholar 

  38. Fan, Z., Yan, Z. & Wen, S. Deep learning and artificial intelligence in sustainability: a review of SDGs, renewable energy, and environmental health. Sustainability 15, 13493 (2023).

    Article  Google Scholar 

  39. Feroz, A. K., Zo, H. & Chiravuri, A. Digital transformation and environmental sustainability: a review and research agenda. Sustainability 13, 1530 (2021).

    Article  Google Scholar 

  40. Cao, X., Liu, L., Cheng, Y. & Shen, X. Towards energy-efficient wireless networking in the big data era: a survey. IEEE Commun. Surv. Tutor. 20, 303–332 (2018).

    Article  Google Scholar 

  41. Freitag, C. et al. The climate impact of ICT: a review of estimates, trends and regulations. Preprint at http://arxiv.org/abs/2102.02622 (2021).

  42. Dhar, P. The carbon impact of artificial intelligence. Nat. Mach. Intell. 2, 423–425 (2020). This article discusses the dual impact of AI on the environment, highlighting how AI can both contribute to carbon emissions, especially during model training, and help mitigate climate change. The article emphasizes the need for transparent quantification of AI’s carbon footprint, proposes strategies for reducing emissions through more efficient AI models and infrastructure, and advocates for policies that promote sustainable AI practices.

    Article  Google Scholar 

  43. Hussain, F., Hassan, S. A., Hussain, R. & Hossain, E. Machine learning for resource management in cellular and IoT networks: potentials, current solutions, and open challenges. IEEE Commun. Surv. Tutor. 22, 1251–1275 (2020).

    Article  Google Scholar 

  44. Chen, M., Miao, Y., Gharavi, H., Hu, L. & Humar, I. Intelligent traffic adaptive resource allocation for edge computing-based 5G networks. IEEE Trans. Cogn. Commun. Netw. 6, 499–508 (2019).

    Article  Google Scholar 

  45. Srivastava, V., Tripathi, S., Singh, K. & Son, L. H. Energy efficient optimized rate based congestion control routing in wireless sensor network. J. Ambient. Intell. Hum. Comput. 11, 1325–1338 (2020).

    Article  Google Scholar 

  46. Serradilla, O., Zugasti, E., Rodriguez, J. & Zurutuza, U. Deep learning models for predictive maintenance: a survey, comparison, challenges and prospects. Appl. Intell. 52, 10934–10964 (2022).

    Article  Google Scholar 

  47. Katurde, A. D. et al. SecureSense: AI/ML based anomaly detection tool. In 2024 Int. Conf. Intelligent Systems for Cybersecurity (ISCS) (eds Chhikara, R. et al.) 1–6 (IEEE, 2024).

  48. Goswami, P. et al. AI based energy efficient routing protocol for intelligent transportation system. IEEE Trans. Intell. Transp. Syst. 23, 1670–1679 (2021).

    Article  Google Scholar 

  49. Ehsan, S. & Hamdaoui, B. A survey on energy-efficient routing techniques with QoS assurances for wireless multimedia sensor networks. IEEE Commun. Surv. Tutor. 14, 265–278 (2012).

    Article  Google Scholar 

  50. Jiang, M. et al. Integrated demand response modeling and optimization technologies supporting energy Internet. Renew. Sust. Energ. Rev. 203, 114757 (2024).

    Article  Google Scholar 

  51. Basit, M. A., Dilshad, S., Badar, R. & Sami Ur Rehman, S. M. Limitations, challenges, and solution approaches in grid-connected renewable energy systems. Int. J. Energy Res. 44, 4132–4162 (2020).

    Article  Google Scholar 

  52. Zhou, Z. et al. Carbon-aware load balancing for geo-distributed cloud services. In 2013 IEEE 21st Int. Symp. Modelling, Analysis and Simulation of Computer and Telecommunication Systems (eds Riley, G. & Walrand, J.) 232–241 (IEEE, 2013).

  53. Ahmad, T., Madonski, R., Zhang, D., Huang, C. & Mujeeb, A. Data-driven probabilistic machine learning in sustainable smart energy/smart energy systems: key developments, challenges, and future research opportunities in the context of smart grid paradigm. Renew. Sust. Energ. Rev. 160, 112128 (2022).

    Article  Google Scholar 

  54. Motepe, S., Hasan, A. N. & Stopforth, R. Improving load forecasting process for a power distribution network using hybrid AI and deep learning algorithms. IEEE Access. 7, 82584–82598 (2019). This paper proposes a novel hybrid AI and deep learning system for load forecasting in South African power-distribution networks, incorporating fuzzy logic, data preprocessing and weather data to improve accuracy. The comparative study shows that long short-term memory outperforms optimally pruned extreme learning machines and adaptive neuro-fuzzy inference systems in forecasting load, especially when including temperature data, and provides valuable insights for maintenance planning.

    Article  Google Scholar 

  55. Chen, C., Liu, Y., Chen, L. & Zhang, C. Bidirectional spatial–temporal adaptive transformer for urban traffic flow forecasting. IEEE Trans. Neural Netw. Learn. Syst. 34, 6913–6925 (2022).

    Article  Google Scholar 

  56. Engeland, K. et al. Space–time variability of climate variables and intermittent renewable electricity production—a review. Renew. Sust. Energ. Rev. 79, 600–617 (2017).

    Article  Google Scholar 

  57. Hamdi, A. et al. Spatiotemporal data mining: a survey on challenges and open problems. Artif. Intell. Rev. 55, 1441–1488 (2022).

    Article  Google Scholar 

  58. Sun, C. et al. Attention-based graph neural networks: a survey. Artif. Intell. Rev. 56, 2263–2310 (2023).

    Article  Google Scholar 

  59. Kong, Z., Jin, X., Xu, Z. & Zhang, B. Spatio-temporal fusion attention: a novel approach for remaining useful life prediction based on graph neural network. IEEE Trans. Instrum. Meas. 71, 1–12 (2022).

    Google Scholar 

  60. Hsu, W.-J., Spyropoulos, T., Psounis, K. & Helmy, A. Modeling spatial and temporal dependencies of user mobility in wireless mobile networks. IEEE/ACM Trans. Netw. 17, 1564–1577 (2009).

    Article  Google Scholar 

  61. Chen, Z. et al. Knowledge graphs meet multi-modal learning: a comprehensive survey. Preprint at http://arxiv.org/abs/2402.05391 (2024).

  62. Qi, W. et al. A resource-efficient cross-domain sensing method for device-free gesture recognition with federated transfer learning. IEEE Trans. Green. Commun. Netw. 7, 393–400 (2023).

    Article  Google Scholar 

  63. Aimen, A., Sidheekh, S., Ladrecha, B., Ahuja, H. & Krishnan, N. C. Adaptation: blessing or curse for higher-way meta-learning. IEEE Trans. Artif. Intell. 5, 1844–1856 (2023).

    Article  Google Scholar 

  64. Liu, Y. et al. Vertical federated learning: concepts, advances, and challenges. IEEE Trans. Knowl. Data Eng. 36, 3615–3634 (2024).

    Article  Google Scholar 

  65. Shi, Z. et al. Artificial intelligence techniques for stability analysis and control in smart grids: methodologies, applications, challenges and future directions. Appl. Energy 278, 115733 (2020).

    Article  Google Scholar 

  66. Barredo Arrieta, A. et al. Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI. Inf. Fusion. 58, 82–115 (2020).

    Article  Google Scholar 

  67. Alam, M. Challenges of integrating spatiotemporal data with AI/ML models for road traffic congestion prediction. J. Adv. Artif. Intell. 1, 8–14 (2025).

    Google Scholar 

  68. Aouedi, O., Le, V. A., Piamrat, K. & Ji, Y. Deep learning on network traffic prediction: recent advances, analysis, and future directions. ACM Comput. Surv. 57, 1–37 (2025).

    Article  Google Scholar 

  69. Wang, J. et al. Inherent spatiotemporal uncertainty of renewable power in China. Nat. Commun. 14, 5379 (2023).

    Article  Google Scholar 

  70. Li, R., Decocq, B., Barros, A., Fang, Y.-P. & Zeng, Z. Estimating 5G network service resilience against short timescale traffic variation. IEEE Trans. Netw. Serv. Manag. 20, 2230–2243 (2023).

    Article  Google Scholar 

  71. Munikoti, S., Agarwal, D., Das, L., Halappanavar, M. & Natarajan, B. Challenges and opportunities in deep reinforcement learning with graph neural networks: a comprehensive review of algorithms and applications. IEEE Trans. Neural Netw. Learn. Syst. 35, 15051–15071 (2023).

    Article  MathSciNet  Google Scholar 

  72. Xiao, H., Pu, X., Pei, W., Ma, L. & Ma, T. A novel energy management method for networked multi-energy microgrids based on improved DQN. IEEE Trans. Smart Grid 14, 4912–4926 (2023).

    Article  Google Scholar 

  73. Liu, K. ang et al. Reliable PPO-based concurrent multipath transfer for time-sensitive applications. IEEE Trans. Veh. Technol. 72, 13575–13590 (2023).

    Article  Google Scholar 

  74. He, B., Meng, Y. & Tang, L. An off-policy reinforcement learning-based adaptive optimization method for dynamic resource allocation problem. IEEE Trans. Neural Netw. Learn. Syst. 36, 3504–3518 (2023).

    Article  MathSciNet  Google Scholar 

  75. Liu, S., He, M., Wu, Z., Lu, P. & Gu, W. Spatial–temporal graph neural network traffic prediction based load balancing with reinforcement learning in cellular networks. Inf. Fusion. 103, 102079 (2024).

    Article  Google Scholar 

  76. Yun, W.-K. & Yoo, S.-J. Q-learning-based data-aggregation-aware energy-efficient routing protocol for wireless sensor networks. IEEE Access. 9, 10737–10750 (2021).

    Article  Google Scholar 

  77. Ning, Z. & Xie, L. A survey on multi-agent reinforcement learning and its application. J. Autom. Intell. 3, 73–91 (2024).

    Google Scholar 

  78. Jayanetti, A., Halgamuge, S. & Buyya, R. Multi-agent deep reinforcement learning framework for renewable energy-aware workflow scheduling on distributed cloud data centers. IEEE Trans. Parallel Distrib. Syst. 35, 604–615 (2024).

    Article  Google Scholar 

  79. Yurtsever, E., Capito, L., Redmill, K. & Ozgune, U. Integrating deep reinforcement learning with model-based path planners for automated driving. In 2020 IEEE Intelligent Vehicles Symposium (IV) (ed. Morris, B.) 1311–1316 (IEEE, 2020).

  80. Sarker, I. H., Janicke, H., Ferrag, M. A. & Abuadbba, A. Multi-aspect rule-based AI: methods, taxonomy, challenges and directions toward automation, intelligence and transparent cybersecurity modeling for critical infrastructures. Internet Things 25, 101110 (2024).

    Article  Google Scholar 

  81. Erol-Kantarci, M. & Mouftah, H. T. Energy-efficient information and communication infrastructures in the smart grid: a survey on interactions and open issues. IEEE Commun. Surv. Tutor. 17, 179–197 (2014).

    Article  Google Scholar 

  82. Xiong, Z., Luo, B., Wang, B.-C., Xu, X. & Huang, T. Multiobjective battery charging strategy based on deep reinforcement learning. IEEE Trans. Transp. Electrification 10, 6893–6903 (2024).

    Article  Google Scholar 

  83. Korkmaz, J. & Ghajar, R. The modified hybrid multi-objective genetic algorithm and loss sensitivity factor for optimal siting and sizing of PV-based distributed generation in distribution networks. In 2023 IEEE 4th Int. Multidisciplinary Conf. Engineering Technology (IMCET) (ed. Sawaya, H.) 69–74 (IEEE, 2023).

  84. Hu, L., Yang, Y., Tang, Z., He, Y. & Luo, X. FCAN-MOPSO: an improved fuzzy-based graph clustering algorithm for complex networks with multiobjective particle swarm optimization. IEEE Trans. Fuzzy Syst. 31, 3470–3484 (2023).

    Article  Google Scholar 

  85. Yang, L., Li, X., Sun, M. & Sun, C. Hybrid policy-based reinforcement learning of adaptive energy management for the energy transmission-constrained island group. IEEE Trans. Ind. Inform. 19, 10751–10762 (2023).

    Article  Google Scholar 

  86. Shirvani, M. H. An energy-efficient topology-aware virtual machine placement in cloud datacenters: a multi-objective discrete JAYA optimization. Sustain. Comput. Inform. Syst. 38, 100856 (2023).

    Google Scholar 

  87. Zhang, H. et al. An energy consumption optimization strategy for wireless sensor networks via multi-objective algorithm. J. King Saud. Univ. Comput. Inf. Sci. 36, 101919 (2024).

    Article  Google Scholar 

  88. Fan, W., Fan, P. & Long, Y. Joint delay-energy optimization for multi-priority random access in machine-type communications. IEEE Trans. Wirel. Commun. 23, 1416–1431 (2023).

    Article  Google Scholar 

  89. El-Afifi, M. I., Sedhom, B. E., Padmanaban, S. & Eladl, A. A. A review of IoT-enabled smart energy hub systems: rising, applications, challenges, and future prospects. Renew. Energy Focus. 25, 100634 (2024).

    Article  Google Scholar 

  90. Hu, S., Chen, X., Ni, W., Hossain, E. & Wang, X. Distributed machine learning for wireless communication networks: techniques, architectures, and applications. IEEE Commun. Surv. Tutor. 23, 1458–1493 (2021). This survey provides a comprehensive review of distributed machine learning techniques, such as federated learning and distributed reinforcement learning, in the context of wireless communications. The article highlights the unique challenges posed by large-scale, geographically dispersed systems and increasing data privacy concerns.

    Article  Google Scholar 

  91. Du, H. et al. Exploring collaborative distributed diffusion-based AI-generated content (AIGC) in wireless networks. IEEE Netw. 38, 178–186 (2023).

    Article  Google Scholar 

  92. Duan, Q. et al. Combining federated learning and edge computing toward ubiquitous intelligence in 6G network: challenges, recent advances, and future directions. IEEE Commun. Surv. Tutor. 25, 2892–2950 (2023).

    Article  Google Scholar 

  93. Shanmugam, L., Tillu, R. & Tomar, M. Federated learning architecture: design, implementation, and challenges in distributed AI systems. J. Knowl. Learn. Sci. Technol. 2, 371–384 (2023).

    Article  Google Scholar 

  94. Shi, Y., Yang, K., Jiang, T., Zhang, J. & Letaief, K. B. Communication-efficient edge AI: algorithms and systems. IEEE Commun. Surv. Tutor. 22, 2167–2191 (2020).

    Article  Google Scholar 

  95. Shao, Y. et al. Distributed graph neural network training: a survey. ACM Comput. Surv. 56, 1–39 (2024).

    Article  Google Scholar 

  96. Yu, W., Ruan, K., Tang, H. & Huang, J. Routing hypergraph convolutional recurrent network for network traffic prediction. Appl. Intell. 53, 16126–16137 (2023).

    Article  Google Scholar 

  97. Wang, Y., Li, Y., Shi, Q. & Wu, Y.-C. ENGNN: a general edge-update empowered GNN architecture for radio resource management in wireless networks. IEEE Trans. Wirel. Commun. 23, 5330–5344 (2023).

    Article  Google Scholar 

  98. Wu, Y., Dai, H.-N. & Tang, H. Graph neural networks for anomaly detection in industrial Internet of Things. IEEE Internet Things J. 9, 9214–9231 (2021).

    Article  Google Scholar 

  99. Barbieri, L., Kianoush, S., Nicoli, M., Serio, L. & Savazzi, S. A close look at the communication efficiency and the energy footprints of robust federated learning in industrial IoT. IEEE Internet Things J. 12, 15130–15150 (2025).

    Article  Google Scholar 

  100. Wang, Q., Chen, Y., Wong, W.-F. & He, B. Scalable and load-balanced full-graph GNN training on multiple GPUs. IEEE Trans. Knowl. Data Eng. 37, 4239–4253 (2025).

    Google Scholar 

  101. Zhang, P. et al. Towards net-zero carbon emissions in network AI for 6G and beyond. IEEE Commun. Mag. 62, 58–64 (2024). This article addresses the challenges of reducing carbon emissions in the development of 6G networks, despite advancements in energy efficiency. The article introduces a dynamic energy-trading and task-allocation framework to reduce carbon emissions in network AI systems, particularly those utilizing federated edge intelligence.

    Article  Google Scholar 

  102. Wu, C.-J. et al. Sustainable AI: environmental implications, challenges and opportunities. Proc. Mach. Learn. Syst. 4, 795–813 (2022).

    Google Scholar 

  103. Ding, R.-X. et al. Large-scale decision-making: characterization, taxonomy, challenges and future directions from an artificial intelligence and applications perspective. Inf. Fusion. 59, 84–102 (2020).

    Article  Google Scholar 

  104. Yeom, S.-K., Shim, K.-H. & Hwang, J.-H. Toward compact deep neural networks via energy-aware pruning. Preprint at http://arxiv.org/abs/2103.10858 (2021).

  105. Bouza, L., Bugeau, A. & Lannelongue, L. How to estimate carbon footprint when training deep learning models? A guide and review. Environ. Res. Commun. 5, 115014 (2023).

    Article  Google Scholar 

  106. Bolón-Canedo, V., Morán-Fernández, L., Cancela, B. & Alonso-Betanzos, A. A review of green artificial intelligence: towards a more sustainable future. Neurocomputing 59, 128096 (2024).

    Article  Google Scholar 

  107. Appio, F. P., Lima, M. & Paroutis, S. Understanding smart cities: innovation ecosystems, technological advancements, and societal challenges. Technol. Forecast. Soc. Change 142, 1–14 (2019).

    Article  Google Scholar 

  108. Jalby, W. et al. The long and winding road toward efficient high-performance computing. Proc. IEEE 106, 1985–2003 (2018).

    Article  Google Scholar 

  109. Song, A., Chen, D. & Zong, Z. Unveiling the truth: an analysis of the energy and carbon footprint of training an OPT model using DeepSpeed on the H100 GPU. In Proc. 14th Int. Green and Sustainable Computing Conf. (eds Krishna, C. M. & Magno, M.) 36–38 (ACM, 2023).

  110. Desislavov, R., Martínez-Plumed, F. & Hernández-Orallo, J. Trends in AI inference energy consumption: beyond the performance-vs-parameter laws of deep learning. Sustain. Comput. Inform. Syst. 38, 100857 (2023).

    Google Scholar 

  111. Gil, Y. et al. Artificial intelligence for modeling complex systems: taming the complexity of expert models to improve decision making. ACM Trans. Interact. Intell. Syst. 11, 1–49 (2021).

    Article  Google Scholar 

  112. Salehi, S. & Schmeink, A. Data-centric green artificial intelligence: a survey. IEEE Trans. Artif. Intell. 5, 1973–1989 (2023).

    Article  Google Scholar 

  113. Castellanos-Nieves, D. & García-Forte, L. Strategies of automated machine learning for energy sustainability in green artificial intelligence. Appl. Sci. 14, 2076–3417 (2024).

    Article  Google Scholar 

  114. Wang, X. & Zhu, W. Advances in neural architecture search. Natl. Sci. Rev. 11, nwae282 (2024).

    Article  Google Scholar 

  115. Wan, Q., Wang, L., Wang, J., Song, S. L. & Fu, X. NAS-SE: designing a highly-efficient in-situ neural architecture search engine for large-scale deployment. In Proc. 56th Annual IEEE/ACM Int. Symposium on Microarchitecture (ed. Pekhimenko, G.) 756–768 (ACM, 2023).

  116. Iman, M., Arabnia, H. R. & Rasheed, K. A review of deep transfer learning and recent advancements. Technologies 11, 40 (2023).

    Article  Google Scholar 

  117. Gharoun, H., Momenifar, F., Chen, F. & Gandomi, A. Meta-learning approaches for few-shot learning: a survey of recent advances. ACM Comput. Surv. 56, 1–41 (2024).

    Article  Google Scholar 

  118. Chungath, T. T., Nambiar, A. M. & Mittal, A. Transfer learning and few-shot learning based deep neural network models for underwater sonar image classification with a few samples. IEEE J. Ocean. Eng. 49, 294–310 (2023).

    Article  Google Scholar 

  119. Yousefpour, A. et al. Green federated learning. Preprint at http://arxiv.org/abs/2303.14604 (2023).

  120. Savazzi, S., Rampa, V., Kianoush, S. & Bennis, M. An energy and carbon footprint analysis of distributed and federated learning. IEEE Trans. Green. Commun. Netw. 7, 248–264 (2022). This paper presents a novel framework for analysing the energy and carbon footprints of distributed and federated learning methods, comparing vanilla federated learning and decentralized approaches. The study underscores the importance of balancing communication efficiency, learner population size, energy consumption and model accuracy to achieve sustainability in distributed learning systems.

    Article  Google Scholar 

  121. Ye, M., Fang, X., Du, B., Yuen, P. C. & Tao, D. Heterogeneous federated learning: state-of-the-art and research challenges. ACM Comput. Surv. 56, 1–44 (2023).

    Google Scholar 

  122. Su, S., Zhou, Z., Ouyang, T., Zhou, R. & Chen, X. Learning to be green: carbon-aware online control for edge intelligence with colocated learning and inference. In 2023 IEEE 43rd Int. Conf. Distributed Computing Systems (ICDCS) (eds Frieder, O. & Jia, X.-H.) 567–578 (IEEE, 2023).

  123. Gligorea, I. et al. Adaptive learning using artificial intelligence in e-learning: a literature review. Educ. Sci. 13, 1216 (2023).

    Article  Google Scholar 

  124. Lattanzi, E., Contoli, C. & Freschi, V. A study on the energy sustainability of early exit networks for human activity recognition. IEEE Trans. Sustain. Comput. 9, 61–74 (2024).

    Article  Google Scholar 

  125. Bothe, S., Farooq, H., Forgeat, J. & Cyras, K. Time-series prediction using nature-inspired small models and curriculum learning. In 2023 IEEE 34th Annual Int. Symp. Personal, Indoor and Mobile Radio Communications (PIMRC) (eds Tong, W. & Zhu, P.-Y.) 1–6 (IEEE, 2023).

  126. Lazzaro, D. et al. Minimizing energy consumption of deep learning models by energy-aware training. In Int. Conf. Image Analysis and Processing (eds Foresti, L. G. et al.) 515–526 (Springer, 2023).

  127. Verdecchia, R., Sallou, J. & Cruz, L. A systematic review of green AI. Wiley Interdiscip. Rev. Data Min. Knowl. Discov. 13, e1507 (2023).

    Article  Google Scholar 

  128. Lacoste, A., Luccioni, A., Schmidt, V. & Dandres, T. Quantifying the carbon emissions of machine learning. Preprint at http://arxiv.org/abs/1910.09700 (2019).

  129. Luccioni, A., Lacoste, A. & Schmidt, V. Estimating carbon emissions of artificial intelligence [opinion]. IEEE Technol. Soc. Mag. 39, 48–51 (2020).

    Article  Google Scholar 

  130. Patterson, D. et al. Carbon emissions and large neural network training. Preprint at http://arxiv.org/abs/2104.10350 (2021).

  131. Schaefer, C. J., Taheri, P., Horeni, M. & Joshi, S. The hardware impact of quantization and pruning for weights in spiking neural networks. IEEE Trans. Circuits Syst. II-Express Brief. 70, 1789–1793 (2023).

    Google Scholar 

  132. Harma, S. B. et al. Effective interplay between sparsity and quantization: from theory to practice. Preprint at http://arxiv.org/abs/2405.20935 (2024).

  133. Singh, R. & Gill, S. S. Edge AI: a survey. Internet Things Cyber-Physical Syst. 3, 71–92 (2023).

    Article  Google Scholar 

  134. Himeur, Y., Sayed, A., Alsalemi, A., Bensaali, F. & Amira, A. Edge AI for Internet of Energy: challenges and perspectives. Internet Things 25, 101035 (2023).

    Article  Google Scholar 

  135. Bullo, M., Jardak, S., Carnelli, P. & Gündüz, D. Sustainable edge intelligence through energy-aware early exiting. In 2023 IEEE 33rd Int. Workshop on Machine Learning for Signal Processing (MLSP) (eds Adaly, T. & Uncini, A.) 1–6 (IEEE, 2023).

  136. Tan, M. & Le, Q. EfficientNet: Rethinking model scaling for convolutional neural networks. In Int. Conf. Machine Learning (eds Chaudhuri, K. & Salakhutdinov, R.) 6105–6114 (PMLR 2019).

  137. Wang, X., Yu, F., Dou, Z.-Y., Darrell, T. & Gonzalez, J. E. SkipNet: learning dynamic routing in convolutional networks. In Proc. Eur. Conf. Computer Vision (ECCV) (eds Ferrari, V. et al.) 409–424 (Springer, 2018).

  138. Hu, T.-K., Chen, T., Wang, H. & Wang, Z. Triple wins: boosting accuracy, robustness and efficiency together by enabling input-adaptive inference. Preprint at http://arxiv.org/abs/2002.10025 (2020).

  139. Deng, Y., Dai, Q. & Zhang, Z. in Artificial Intelligence, Evolutionary Computing and Metaheuristics: In the Footsteps of Alan Turing (ed. Yang, X.-S.) 345–369 (Springer, 2013).

  140. Dai, S., Genc, H., Venkatesan, R. & Khailany, B. Efficient transformer inference with statically structured sparse attention. In 2023 60th ACM/IEEE Design Automation Conf. (DAC) (ed. Henkel, J.) 1–6 (IEEE, 2023).

  141. Zawish, M., Ashraf, N., Ansari, R. I. & Davy, S. Energy-aware AI-driven framework for edge-computing-based IoT applications. IEEE Internet Things J. 10, 5013–5023 (2022).

    Article  Google Scholar 

  142. Yokoyama, A. M., Ferro, M., de Paula, F. B., Vieira, V. G. & Schulze, B. Investigating hardware and software aspects in the energy consumption of machine learning: a green AI-centric analysis. Concurr. Comput. Pract. Exp. 35, e7825 (2023).

    Article  Google Scholar 

  143. Balaram, V. Rare earth elements: a review of applications, occurrence, exploration, analysis, recycling, and environmental impact. Geosci. Front. 10, 1285–1303 (2019).

    Article  Google Scholar 

  144. Ligozat, A.-L., Lefevre, J., Bugeau, A. & Combaz, J. Unraveling the hidden environmental impacts of AI solutions for environment life cycle assessment of AI solutions. Sustainability 14, 5172 (2022). This article examines the environmental impacts of AI, particularly its role in addressing greenhouse gas emissions, while highlighting the energy and greenhouse gas costs of training large AI models. The article proposes a study of the potential negative environmental effects of AI for green, using life-cycle assessment methodologies to analyse these impacts and evaluate the overall environmental benefits of AI solutions while addressing gaps in current research.

    Article  Google Scholar 

  145. Bernardo, P. P., Gerum, C., Frischknecht, A., Lübeck, K. & Bringmann, O. UltraTrail: a configurable ultralow-power TC-ResNet AI accelerator for efficient keyword spotting. IEEE Trans. Comput-Aided Des. Integr. Circuits Syst. 39, 4240–4251 (2020).

    Article  Google Scholar 

  146. Shiri, A. et al. E2HRL: an energy-efficient hardware accelerator for hierarchical deep reinforcement learning. ACM Transact. Des. Automat. Electron. Syst. TODAES 27, 1–19 (2022).

    Article  Google Scholar 

  147. Chen, Z., Blair, H. T. & Cong, J. Energy-efficient LSTM inference accelerator for real-time causal prediction. ACM Transact. Des. Automat. Electron. Syst. TODAES 27, 1–19 (2022).

    Google Scholar 

  148. Choi, C. et al. Reconfigurable heterogeneous integration using stackable chips with embedded artificial intelligence. Nat. Electron. 5, 386–393 (2022).

    Article  Google Scholar 

  149. Shastri, B. J. et al. Photonics for artificial intelligence and neuromorphic computing. Nat. Photonics 15, 102–114 (2021).

    Article  Google Scholar 

  150. Zhang, X., Jiang, W., Shi, Y. & Hu, J. When neural architecture search meets hardware implementation: from hardware awareness to co-design. In 2019 IEEE Computer Society Annual Symposium on VLSI (ISVLSI) (eds Wen, W.-J. et al.) 25–30 (IEEE, 2019).

  151. Bouzidi, H., Odema, M., Ouarnoughi, H., Al Faruque, M. A. & Niar, S. HADAS: hardware-aware dynamic neural architecture search for edge performance scaling. In 2023 Design, Automation & Test in Europe Conf. Exhibition (DATE) (ed. O’Connor, I.) 1–6 (IEEE, 2023).

  152. Tang, Z., Wang, Y., Wang, Q. & Chu, X. The impact of GPU DVFS on the energy and performance of deep learning: an empirical study. In Proc. Tenth ACM Int. Conf. Future Energy Systems (eds Lin, X.-J. & Low, S.) 315–325 (ACM, 2019).

  153. Drechsler, R., Metz, C. A. & Plump, C. Energy-efficient CNN inferencing on GPUs with dynamic frequency scaling. In Int. Conf. Innovations in Data Analytics (eds Bhattacharya, A. et al.) 375–389 (Springer, 2023).

  154. Rao, A., Plank, P., Wild, A. & Maass, W. A long short-term memory for AI applications in spike-based neuromorphic hardware. Nat. Mach. Intell. 4, 467–479 (2022).

    Article  Google Scholar 

  155. Shrestha, A. et al. A survey on neuromorphic computing: models and hardware. IEEE Circuits Syst. Mag. 22, 6–35 (2022).

    Article  Google Scholar 

  156. Akopyan, F. et al. TrueNorth: design and tool flow of a 65 mW 1 million neuron programmable neurosynaptic chip. IEEE Trans. Comput-Aided Des. Integr. Circuits Syst. 34, 1537–1557 (2015).

    Article  Google Scholar 

  157. Ganguly, C., Meem, M. Z., Kabir, S. K. & Biswas, S. N. Analysis of a low-power full adder and half adder using a new adiabatic logic. In 2023 26th Int. Conf. Computer and Information Technology (ICCIT) (ed. Rahardja, U.) 1–5 (IEEE, 2023).

  158. Jiang, L. & Chen, F. CarbonScaling: extending neural scaling laws for carbon footprint in large language models. Preprint at https://arxiv.org/abs/2508.06524 (2025).

  159. Wang, X., Du, H., Gao, Y. & Kim, D. I. AOLO: analysis and optimization for low-carbon oriented wireless large language model services. Preprint at https://arxiv.org/abs/2503.04418 (2025).

  160. International Energy Agency. Energy and AI. IEA https://www.iea.org/reports/energy-and-ai (2025).

  161. Fu, Z., Chen, F., Zhou, S., Li, H. & Jiang, L. LLMCO2: advancing accurate carbon footprint prediction for LLM inferences. ACM SIGENERGY Energy Inform. Rev. 5, 63–68 (2025).

    Article  Google Scholar 

  162. Režun, T. & Kapusta, D. The energy and water footprint of generative AI: a vanguard leadership perspective. Int. Leadersh. J. 17, 60–67 (2025).

    Google Scholar 

  163. Jegham, N., Abdelatti, M., Elmoubarki, L. & Hendawi, A. How hungry is AI? Benchmarking energy, water, and carbon footprint of LLM inference. Preprint at https://arxiv.org/abs/2505.09598 (2025).

  164. Xu, X. et al. Scaling for edge inference of deep neural networks. Nat. Electron. 1, 216–222 (2018).

    Article  Google Scholar 

  165. Cao, Z., Zhou, X., Hu, H., Wang, Z. & Wen, Y. Toward a systematic survey for carbon neutral data centers. IEEE Commun. Surv. Tutor. 24, 895–936 (2022).

    Article  Google Scholar 

  166. Khan, T., Tian, W., Ilager, S. & Buyya, R. Workload forecasting and energy state estimation in cloud data centres: ML-centric approach. Futur. Gener. Comp. Syst. 128, 320–332 (2022).

    Article  Google Scholar 

  167. Cui, Y., Cao, K. & Wei, T. Reinforcement learning-based device scheduling for renewable energy-powered federated learning. IEEE Trans. Industr. Inform. 19, 6264–6274 (2022).

    Article  Google Scholar 

  168. Lin, T., Stich, S. U., Barba, L., Dmitriev, D. & Jaggi, M. Dynamic model pruning with feedback. Preprint at https://arxiv.org/abs/2006.07253 (2020).

  169. Elthakeb, A. T., Pilligundla, P., Mireshghallah, F., Yazdanbakhsh, A. & Esmaeilzadeh, H. ReLeQ: a reinforcement learning approach for automatic deep quantization of neural networks. IEEE Micro 40, 37–45 (2020).

    Article  Google Scholar 

  170. Gou, J., Yu, B., Maybank, S. J. & Tao, D. Knowledge distillation: a survey. Int. J. Comput. Vis. 129, 1789–1819 (2021).

    Article  Google Scholar 

  171. Xi, L. Optimizing product lifecycle management with AI: from development to deployment. Int. IT J. Res. 2, 8–14 (2024).

    Google Scholar 

  172. Forsberg, E. & Harris, C. Teaching an AI to Recycle by Looking at Scrap Metal: Semantic Segmentation through Self-Supervised Learning with Transformers. MSc thesis, Linköping Univ. (2022).

  173. Midgley M. Practical changes could reduce AI energy demand by up to 90%. UCL News https://www.ucl.ac.uk/news/2025/jul/practical-changes-could-reduce-ai-energy-demand-90 (2025).

  174. Li, Y., Tolosa, L., Rivas-Echeverria, F. & Marquez, R. Integrating AI in education: navigating UNESCO global guidelines, emerging trends, and its intersection with sustainable development goals. Preprint at https://doi.org/10.26434/chemrxiv-2025-wz4n9 (2025).

  175. Zewe, A. Explained: generative AI’s environmental impact. MIT News https://news.mit.edu/2025/explained-generative-ai-environmental-impact-0117 (2025).

  176. Lechowicz, A. et al. Carbon- and precedence-aware scheduling for data processing clusters. In ACM Special Interest Group on Data Communication 2025 (eds Curado, M. & Rothenberg, C. E.) 1241–1244 (ACM, 2025).

  177. Gu, D. et al. GreenFlow: a carbon-efficient scheduler for deep learning workloads. IEEE Trans. Parallel Distrib. Syst. 36, 168–184 (2025).

    Article  Google Scholar 

  178. Sorrell, S. Jevons’ paradox revisited: the evidence for backfire from improved energy efficiency. Energy Policy 37, 1456–1469 (2009).

    Article  Google Scholar 

  179. Alkurd, R., Abualhaol, I. Y. & Yanikomeroglu, H. Personalized resource allocation in wireless networks: an AI-enabled and big data-driven multi-objective optimization. IEEE Access. 8, 144592–144609 (2020).

    Article  Google Scholar 

  180. Mazidi, M. R., Aghazadeh, M., Teshnizi, Y. A. & Mohagheghi, E. Optimal placement of switching devices in distribution networks using multi-objective genetic algorithm NSGAII. In 2013 21st Iranian Conf. Electrical Engineering (ICEE) (eds Almasgna, F. & Akbari, A.) 1–6 (IEEE, 2013).

  181. Hosna, A. et al. Transfer learning: a friendly introduction. J. Big Data 9, 102 (2022).

    Article  Google Scholar 

  182. Wang, Y., Yao, Q., Kwok, J. T. & Ni, L. M. Generalizing from a few examples: a survey on few-shot learning. ACM Comput. Surv. 53, 63 (2020).

    Google Scholar 

  183. Fontenla-Romero, O., Guijarro-Berdiñas, B., Hernández-Pereira, E. & Pérez-Sánchez, B. An effective and efficient green federated learning method for one-layer neural networks. In Proc. 39th ACM/SIGAPP Symposium on Applied Computing (ed. Hong, J.-M.) 1050–1052 (ACM, 2024).

  184. Han, S., Pool, J., Tran, J. & Dally, W. Learning both weights and connections for efficient neural network. Adv. Neural Inf. Process. Syst. 28, 1135–1143 (2015).

    Google Scholar 

  185. Han, Y. et al. Dynamic neural networks: a survey. IEEE Trans. Pattern Anal. Mach. Intell. 44, 7436–7456 (2021).

    Article  Google Scholar 

  186. Mao, H. et al. Exploring the regularity of sparse structure in convolutional neural networks. Preprint at http://arxiv.org/abs/1705.08922 (2017).

  187. Matsumoto, K., Mori, H. & Orii, Y. Cooling from the bottom side (laminate (substrate) side) of a three-dimensional (3D) chip stack. In 2014 Int. 3D Systems Integration Conf. (3DIC) (eds Franzon, P. & Garrou, P.) 1–6 (IEEE, 2014).

  188. Chitty-Venkata, K. T., Bian, Y., Emani, M., Vishwanath, V. & Somani, A. K. Differentiable neural architecture, mixed precision and accelerator co-search. IEEE Access. 11, 106670–106687 (2023).

    Article  Google Scholar 

  189. Wang, Y., Wang, Z., Xing, N. & Zhao, S. UAV Coverage path planning based on deep reinforcement learning. In 2023 IEEE 6th Int. Conf. Computer and Communication Engineering Technology (CCET) (ed. Ma, L.) 143–147 (IEEE, 2023).

  190. Yang, W. & Thapliyal, H. Low-power and energy-efficient full adders with approximate adiabatic logic for edge computing. In 2020 IEEE Computer Society Annual Symposium on VLSI (ISVLSI) (eds Theocharides, T. & Narayanan, V.) 312–315 (IEEE, 2020).

  191. Ohalete, N. C. et al. AI-driven solutions in renewable energy: a review of data science applications in solar and wind energy optimization. World J. Adv. Res. Rev. 20, 401–417 (2023).

    Article  Google Scholar 

Download references

Acknowledgements

This work was supported in part by the National Key R&D Program of China (grants 2024YFE0200500 and 2024YFE0200504); the National Natural Science Foundation of China Major International Joint Research Project (grant 62120106007); the Hubei Province international joint research project (grant 2024EHA036); and the Interdisciplinary Research Program of Huazhong University of Science and Technology (grant 2024JCYJ022).

Author information

Authors and Affiliations

  1. School of Electronic Information and Communications, Huazhong University of Science and Technology, Wuhan, China

    Junliang Ye, Yuxi Zhao, Yue Yu, Yuting Li & Xiaohu Ge

  2. Gharavi Consulting Services, Gaithersburg, MD, USA

    Hamid Gharavi

  3. Shanghai Center of the Hong Kong University of Science and Technology, Shanghai, China

    Yang Yang

  4. Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai, China

    Wuxiong Zhang

Authors
  1. Junliang Ye
    View author publications

    Search author on:PubMed Google Scholar

  2. Yuxi Zhao
    View author publications

    Search author on:PubMed Google Scholar

  3. Yue Yu
    View author publications

    Search author on:PubMed Google Scholar

  4. Yuting Li
    View author publications

    Search author on:PubMed Google Scholar

  5. Xiaohu Ge
    View author publications

    Search author on:PubMed Google Scholar

  6. Hamid Gharavi
    View author publications

    Search author on:PubMed Google Scholar

  7. Yang Yang
    View author publications

    Search author on:PubMed Google Scholar

  8. Wuxiong Zhang
    View author publications

    Search author on:PubMed Google Scholar

Contributions

J.Y., Y.Z., Y. Yu, Y.L. and X.G. researched data for the article. All authors contributed substantially to discussion of the content. J.Y., Y.Z., Y. Yu, Y.L., X.G., Y. Yang and W.Z. wrote the article. All authors edited the manuscript before submission.

Corresponding author

Correspondence to Xiaohu Ge.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Reviews Electrical Engineering thanks Alok Mishra, Loveleen Gaur and the other anonymous reviewers for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Glossary

Deep learning

A type of machine learning that uses artificial neural networks with multiple layers (‘deep’) to learn complex patterns from vast amounts of data.

Deep Q-networks

(DQNs). A deep reinforcement learning method that uses a neural network to approximate the action-value function (Q-function), employing experience replay and target networks for stable training.

Deep RL

A method that combines deep neural networks and reinforcement learning (RL) to solve complex, dynamic control problems in high-dimensional state spaces.

Edge AI

A paradigm that moves artificial intelligence (AI) computation from cloud data centres to edge devices closer to the data source, reducing energy consumption and latency by localizing processing.

Federated learning

A distributed, privacy-preserving training paradigm in which models are trained across multiple decentralized devices without raw data leaving the local device.

Few-shot learning

A technique enabling models to generalize new tasks or categories from only a few labelled examples.

Graph neural networks

(GNNs). A class of deep learning models designed to operate on graph-structured data, capturing spatial dependencies and topological structures between nodes and edges.

Machine learning

A class of data-driven methods that learn statistical models or decision policies from observations, enabling prediction and control without explicit programming.

Meta-learning

‘Learning to learn’ methods that train models to rapidly adapt to new tasks using only limited additional data.

Model compression

A technique to reduce the size and computational complexity of trained artificial intelligence models, including quantization, pruning and knowledge distillation.

Multimodal learning

Learning from and integrating multiple data modalities (for example, text, images and sensor signals) to improve prediction or decision-making.

Multi-objective optimization

An optimization approach that seeks to simultaneously balance multiple conflicting objectives to find optimal trade-offs.

Neural architecture search

Automated methods that search for neural network architectures meeting desired objectives such as accuracy, latency, memory or energy constraints.

Neuromorphic computing

A brain-inspired computing paradigm in which chips mimic biological neurons and synapses to perform artificial intelligence computations at extremely low power.

Proximal policy optimization

A policy-gradient deep reinforcement learning algorithm that uses a clipped objective function to limit policy updates, improving training stability and sample efficiency.

Reinforcement learning

(RL). An artificial intelligence approach in which an agent learns to optimize its strategy by interacting with an environment, taking actions and receiving rewards or penalties.

Transfer learning

A technique that applies knowledge learned from a source task or domain to a different but related target task, reducing data and training requirements.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ye, J., Zhao, Y., Yu, Y. et al. Artificial intelligence for low-carbon energy and information networks. Nat Rev Electr Eng 3, 238–253 (2026). https://doi.org/10.1038/s44287-026-00271-0

Download citation

  • Accepted:

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1038/s44287-026-00271-0

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing