A collective AI via lifelong learning and sharing at the edge
Open accessThe DOI matches its Crossref record; 181 of the 181 references it cites were found in a registry or an index.
Abstract
One vision of a future artificial intelligence (AI) is where many separate units can learn independently over a lifetime and share their knowledge with each other. The synergy between lifelong learning and sharing has the potential to create a society of AI systems, as each individual unit can contribute to and benefit from the collective knowledge. Essential to this vision are the abilities to learn multiple skills incrementally during a lifetime, to exchange knowledge among units via a common language, to use both local data and communication to learn, and to rely on edge devices to host the necessary decentralized computation and data. The result is a network of agents that can quickly respond to and learn new tasks, that collectively hold more knowledge than a single agent and that can extend current knowledge in more diverse ways than a single agent. Open research questions include when and what knowledge should be shared to maximize both the rate of learning and the long-term learning performance. Here we review recent machine learning advances converging towards creating a collective machine-learned intelligence. We propose that the convergence of such scientific and technological advances will lead to the emergence of new types of scalable, resilient and sustainable AI systems. An emerging research area in AI is developing multi-agent capabilities with collections of interacting AI systems. Andrea Soltoggio and colleagues develop a vision for combining such approaches with current edge computing technology and lifelong learning advances. The envisioned network of AI agents could quickly learn new tasks in open-ended applications, with individual AI agents independently learning and contributing to and benefiting from collective knowledge.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
Boyd, S., Parikh, N., Chu, E., Peleato, B., & Eckstein, J. (2011) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000016 · Cited by 17,075
Edge computing: Vision and challenges
Shi, W., Cao, J., Zhang, Q., Li, Y., & Xu, L. (2016) · IEEE Internet of Things Journal · doi:10.1109/jiot.2016.2579198 · Cited by 8,733 · cited 2×
“Category 1 is typical of edge-AI devices that incorporate trained models for inference, for example, smartphones, smart cameras and autonomous vehicles 17,113,114 .”
Overcoming catastrophic forgetting in neural networks
Kirkpatrick, J., Pascanu, R., Rabinowitz, N., Veness, J., Desjardins, G., Rusu, A. A., Milan, K., Quan, J., Ramalho, T., Grabska-Barwinska, A., et al. (2017) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1611835114 · Cited by 7,899
Advances and Open Problems in Federated Learning
Kairouz, P., & McMahan, H. B. (2021) · Foundations and Trends in Machine Learning · doi:10.1561/2200000083 · Cited by 7,772
Intelligent agents: theory and practice
Wooldridge, M., & Jennings, N. R. (1995) · The Knowledge Engineering Review · doi:10.1017/s0269888900008122 · Cited by 6,810
A Comprehensive Survey on Transfer Learning
Zhuang, F., Qi, Z., Duan, K., Xi, D., Zhu, Y., Zhu, H., Xiong, H., & He, Q. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2020.3004555 · Cited by 6,678 · cited 1×
“Reusing previously acquired knowledge when learning a new task is key to transfer learning 62–64 .”
Caruana, R. (1997) · Machine Learning · doi:10.1023/a:1007379606734 · Cited by 6,636
International Conference on Learning Representations (ICLR 2013)
松尾, 豊. (2013) · 人工知能学会誌 = Journal of Japanese Society for Artificial Intelligence · Cited by 6,286
A Survey on Mobile Edge Computing: The Communication Perspective
Mao, Y., You, C., Zhang, J., Huang, K., & Letaief, K. B. (2017) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2017.2745201 · Cited by 5,607 · cited 1×
“Category 1 is typical of edge-AI devices that incorporate trained models for inference, for example, smartphones, smart cameras and autonomous vehicles 17,113,114 .”
Research works printed but not linked: 31
- Ferber, J. & Weiss, G. Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence Vol. 1 (Addison-Wesley, 1999).
- Mendez, J. A., van Seijen, H. & Eaton, E. Modular lifelong reinforcement learning via neural composition. In International Conference on Learning Representations (2022).
- Li, T., Sahu, A. K., Talwalkar, A. & Smith, V. Federated learning: challenges, methods, and future directions. IEEE Signal Process. Mag. 37, 50–60 (2020).
- Zhang, J., Cheung, B., Finn, C., Levine, S. & Jayaraman, D. Cautious adaptation for reinforcement learning in safety-critical settings. In International Conference on Machine Learning 11055–11065 (PMLR, 2020).
- McMahan, B., Moore, E., Ramage, D., Hampson, S. & Arcas, B. A. Y. Communication-efficient learning of deep networks from decentralized data. In Artificial Intelligence and Statistics 1273–1282, (PMLR, 2017).
- Silver, D. L., Yang, Q. & Li, L. Lifelong machine learning systems: beyond learning algorithms. In 2013 AAAI Spring Symposium Series (AAAI, 2013).
- Mendez, J. A. & Eaton, E. Lifelong learning of compositional structures. In International Conference on Learning Representations (2021).
- Xie, A. & Finn, C. Lifelong robotic reinforcement learning by retaining experiences. In Conference on Lifelong Learning Agents 838–855 (PMLR, 2022).
- Ben-Iwhiwhu, E., Nath, S., Pilly, P. K., Kolouri, S. & Soltoggio, A. Lifelong reinforcement learning with modulating masks. In Transactions on Machine Learning Research (2023).
- Tasse, G. N., James, S. & Rosman, B. Generalisation in lifelong reinforcement learning through logical composition. In International Conference on Learning Representations (2022).
- De Lange, M., van de Ven, G. & Tuytelaars, T. Continual evaluation for lifelong learning: identifying the stability gap. In 11th International Conference on Learning Representations https://openreview.net/forum?id=Zy350cRstc6 (ICLR, 2023).
- Yoon, J., Jeong, W., Lee, G., Yang, E. & Hwang, S. J. Federated continual learning with weighted inter-client transfer. In International Conference on Machine Learning 12073–12086 (PMLR, 2021).
- Taylor, M. E. & Stone, P. Transfer learning for reinforcement learning domains: a survey. J. Mach. Learn. Res. 10, 1633–1685 (2009).
- Koohpayegani, S. A., Navaneet, K., Nooralinejad, P., Kolouri, S. & Pirsiavash, H. NOLA: networks as linear combination of low rank random basis. In International Conference on Learning Representations (ICLR, 2024).
- Musliner, D. J. et al. OpenMIND: planning and adapting in domains with novelty. In Proc. 9th Conference on Advances in Cognitive Systems (2021).
- Mendez, J. & Eaton, E. How to reuse and compose knowledge for a lifetime of tasks: a survey on continual learning and functional composition. In Transactions on Machine Learning Research (2023).
- Zoph, B. & Le, Q. Neural architecture search with reinforcement learning. In International Conference on Learning Representations (2017).
- Lin, J. et al. On-device training under 256kb memory. 36th Conference on Neural Information Processing Systems (NeurIPS)(2022).
- Hayes, T. L. & Kanan, C. Online continual learning for embedded devices. In Proc. First Conference on Lifelong Learning Agents (eds Chandar, S. et al.) 744–766 (PMLR, 2022).
- Rostami, M., Kolouri, S., Kim, K. & Eaton, E. Multi-agent distributed lifelong learning for collective knowledge acquisition. In Proc. 17th International Conference on Autonomous Agents and Multiagent Systems 2018 (2017).
- Koster, N., Grothe, O. & Rettinger, A. Signing the supermask: keep, hide, invert. In International Conference on Learning Representations https://openreview.net/forum?id=e0jtGTfPihs (2022).
- Saha, G., Garg, I. & Roy, K. Gradient projection memory for continual learning. In International Conference on Learning Representations (2021).
- Singh, P., Verma, V. K., Mazumder, P., Carin, L. & Rai, P. Calibrating CNNs for lifelong learning. Adv. Neural Inf. Process. Syst. 33, 15579–15590 (2020).
- Ma, Z., Lu, Y., Li, W. & Cui, S. EFL: elastic federated learning on non-IID data. In Conference on Lifelong Learning Agents 92–115 (PMLR, 2022).
- Baradad Jurjo, M., Wulff, J., Wang, T., Isola, P. & Torralba, A. Learning to see by looking at noise. Adv. Neural Inf. Process. Syst. 34, 2556–2569 (2021).
- Zheng, G., Zhou, S., Braverman, V., Jacobs, M. A. & Parekh, V. S. Selective experience replay compression using coresets for lifelong deep reinforcement learning in medical imaging. In Proc. Machine Learning Research 227, 1751–1764 (2024).
- Shperberg, S. S., Liu, B., Allievi, A. & Stone, P. A rule-based shield: Accumulating safety rules from catastrophic action effects. In Conference on Lifelong Learning Agents 231–242 (PMLR, 2022).
- Garcıa, J. & Fernández, F. A comprehensive survey on safe reinforcement learning. J. Mach. Learn. Res. 16, 1437–1480 (2015).
- Sitzmann, V., Martel, J., Bergman, A., Lindell, D. & Wetzstein, G. Implicit neural representations with periodic activation functions. Adv. Neural Inf. Process. Syst. 33, 7462–7473 (2020).
- Bylow, E., Sturm, J., Kerl, C., Kahl, F. & Cremers, D. Real-time camera tracking and 3D reconstruction using signed distance functions. Rob. Sci. Syst. 2, 2 (2013).
- Kozik, R., Choras, M. & Keller, J. Balanced efficient lifelong learning (B-ELLA) for cyber attack detection. J. Univers. Comput. Sci. 25, 2–15 (2019).
Cited by
A survey on knowledge distillation: Recent advancements
Moslemi, A., Briskina, A., Dang, Z., & Li, J. H. (2024) · Machine Learning with Applications · doi:10.1016/j.mlwa.2024.100605 · Cited by 235
Upadhyay, R., Phlypo, R., Saini, R., & Liwicki, M. (2024) · IEEE Access · doi:10.1109/access.2024.3478805 · Cited by 55
Scaling Up Multi-Agent Reinforcement Learning: An Extensive Survey on Scalability Issues
Liu, D., Ren, F., Yan, J. H., Su, G., Gu, W., & Kato, S. (2024) · IEEE Access · doi:10.1109/access.2024.3410318 · Cited by 64
Designing AI-powered learning: adult learners’ expectations for curriculum and human-AI interaction
Kim, J., Yu, S., Detrick, R., Lin, X. Y., & Li, N. (2025) · Educational Technology Research and Development · doi:10.1007/s11423-025-10549-z · Cited by 47
Ji, Z., & Jiang, M. (2025) · Renewable and Sustainable Energy Reviews · doi:10.1016/j.rser.2025.116159 · Cited by 53
Metadata
- Authors
- Soltoggio, A., Ben-Iwhiwhu, E., Braverman, V., Eaton, E., Epstein, B., Ge, Y., Halperin, L., How, J., Itti, L., Jacobs, M. A., Kantharaju, P., Le, L., Lee, S., Liu, X., Monteiro, S. T., Musliner, D., Nath, S., Panda, P., Peridis, C., Pirsiavash, H., Parekh, V., Roy, K., Shperberg, S., Siegelmann, H. T., Stone, P., Vedder, K., Wu, J., Yang, L., Zheng, G., & Kolouri, S.
- Affiliations
- Loughborough University (GB)
- Rice University (US)
- University of Pennsylvania (US)
- ECS (United States) (US)
- University of Southern California (US)
- Massachusetts Institute of Technology (US)
- Johns Hopkins University (US)
- Smart Information Flow Technologies (United States) (US)
- Aurora Flight Sciences (United States) (US)
- Vanderbilt University (US)
- Yale University (US)
- University of California, Davis (US)
- University of Maryland, Baltimore (US)
- Purdue University West Lafayette (US)
- Ben-Gurion University of the Negev (IL)
- University of Massachusetts Amherst (US)
- The University of Texas at Austin (US)
- University of California, Berkeley (US)
- University of California, Los Angeles (US)
- Corresponding author
- Soltoggio, Andrea
- ORCID
- Soltoggio, Andrea
- Ben-Iwhiwhu, Eseoghene
- Braverman, Vladimir
- Eaton, Eric
- Epstein, Benjamin
- Ge, Yunhao
- How, Jonathan
- Itti, Laurent
- Jacobs, Michael A.
- Kantharaju, Pavan
- Le, Long
- Lee, Steven
- Liu, Xinran
- Monteiro, Sildomar T.
- Musliner, David
- Nath, Saptarshi
- Panda, Priyadarshini
- Pirsiavash, Hamed
- Parekh, Vishwa
- Roy, Kaushik
- Shperberg, Shahaf
- Siegelmann, Hava T.
- Stone, Peter
- Wu, Jingfeng
- Yang, Lin
- Kolouri, Soheil
- Venue
- Nature Machine Intelligence
- Published
- 2024-03-22
- Volume
- 6
- Issue
- 3
- Pages
- 251-264
- ISSN
- 2522-5839 (ISSN-L)
- OpenAlex
- W4393097270
- Cited by
- 51
- References
- 173
- Citations by year
- 2026: 18 · 2025: 26 · 2024: 7
- FWCI
- 6.60
- Type
- journal article
- Language
- en
- Open access
- green
- figshare.com
- Licence
- cc-by
- www.springernature.com/gp/researchers/text-and-data-mining (tdm)
- www.springernature.com/gp/researchers/text-and-data-mining (vor)
- Locations
- Nature Machine Intelligence (publishedVersion)
- Figshare (submittedVersion, cc-by)
- Funders
- United States Department of Defense | Defense Advanced Research Projects Agency (HR00112190132, HR00112190135, HR00112190130, HR00112190133, HR00112190134, HR00112190136)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Keywords
- lifelong learning, knowledge sharing, edge computing, multi-agent systems, decentralized learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: A collective AI via lifelong learning and sharing at the edge. https://app.cochat.ai/papers/10.1038/s42256-024-00800-2
- The paper
- Soltoggio, A., Ben-Iwhiwhu, E., Braverman, V., Eaton, E., Epstein, B., Ge, Y., Halperin, L., How, J., Itti, L., Jacobs, M. A., Kantharaju, P., Le, L., Lee, S., Liu, X., Monteiro, S. T., Musliner, D., Nath, S., Panda, P., Peridis, C., . . . Kolouri, S. (2024). A collective AI via lifelong learning and sharing at the edge. Nature Machine Intelligence. https://doi.org/10.1038/s42256-024-00800-2
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.