An official website of the United States government
Drawing on parallels with biological immunity, this paper introduces a new use case for learning at the edge called survival-critical machine learning (SCML). Unlike federated learning, which assumes supervised learning with pre-labeled data, SCML involves semi-supervised learning in streaming settings where labels may need to be obtained at very low network bandwidth and extreme class imbalance. We show that the recently-developed workflow of Live Learning is a good fit for SCML. Starting from a weak bootstrap model, this workflow seamlessly pipelines semi-supervised learning, active learning, and transfer learning, with asynchronous bandwidth-sensitive data transmission for labeling. As improved models evolve at the edge through periodic re-training, the threat detection ability of the SCML system improves. This, in turn, improves the survivability of the host system.
more »
« less
Active Learning for Optimal Experimental Design in Machine Learning-Based Building Energy System Identification
Code for preprint on arXiv in 2026/06: "Active Learning for Optimal Experimental Design in Machine Learning-Based Building Energy System Identification"
more »
« less
- PAR ID:
- 10693513
- Publisher / Repository:
- Zenodo
- Date Published:
- Edition / Version:
- v1.0.0
- Subject(s) / Keyword(s):
- Active learning HVAC Optimal Experiment Design Machine Learning
- Format(s):
- Medium: X
- Right(s):
- MIT License
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The primary goal of the authentic learning approach is to engage and motivate students in learning real world problem solving. We report our experience in developing k-nearest neighbor (KNN) classification for anomaly user behavior detection, one of the authentic machine learning for cybersecurity (ML4Cybr) learning modules based on 10 cybersecurity (CybrS) cases with machine learning (ML) solutions. All portable labs are made available on Google CoLab. So students can access and practice these hands-on labs anywhere and anytime without software installation and configuration which will engage students in learning concepts immediately and getting more experience for hands-on problem solving skills.more » « less
-
Tang, Pingbo; Grau, David; Asmar, Mounir E. (Ed.)Effective construction engineering and management education requires hands-on experiences that have not traditionally been offered in classroom settings. Physical building competitions like Solar Decathlon are valuable for providing experiential learning opportunities that may support tacit and explicit knowledge development among students, but they are often not available to all students due to funding and resource limitations. Less resource intensive teaching strategies, such as project based learning, can mimic the benefit of physical experiences by providing context to learning content. This paper reviews project based learning literature to identify trends in reported learning gains from the adoption of this strategy. Additionally, emerging technologies offer the ability to create low cost, immersive multimedia environments that may be able to support the types of learning targeted by physical design and construction experiences. Literature on multimedia learning theory is explored to identify opportunities for multimedia applications to facilitate learnings derived by physical educational contexts, but with the use of increasingly affordable multimedia strategies. This paper resulted in identifying six learning gains that have a theoretical potential to be facilitated using augmented reality and virtual reality technologies. The theoretical potential was deduced based on prior research on teaching strategies that provide real-world context to learning content. The authors of this paper propose using the identified learning gains as targets to specifically design implementation studies to verify this potential. The learning gains identified in the results section can be targeted and measured in future research when empirically validating the use of immersive technologies for construction education. The contribution of this work is in synthesizing the learning gains that future researchers should target based on evidence from prior research in related learning contexts.more » « less
-
Effective construction engineering and management education requires hands-on experiences that have not traditionally been offered in classroom settings. Physical building competitions like Solar Decathlon are valuable for providing experiential learning opportunities that may support tacit and explicit knowledge development among students, but they are often not available to all students due to funding and resource limitations. Less resource intensive teaching strategies, such as project based learning, can mimic the benefit of physical experiences by providing context to learning content. This paper reviews project based learning literature to identify trends in reported learning gains from the adoption of this strategy. Additionally, emerging technologies offer the ability to create low cost, immersive multimedia environments that may be able to support the types of learning targeted by physical design and construction experiences. Literature on multimedia learning theory is explored to identify opportunities for multimedia applications to facilitate learnings derived by physical educational contexts, but with the use of increasingly affordable multimedia strategies. This paper resulted in identifying six learning gains that have a theoretical potential to be facilitated using augmented reality and virtual reality technologies. The theoretical potential was deduced based on prior research on teaching strategies that provide real-world context to learning content. The authors of this paper propose using the identified learning gains as targets to specifically design implementation studies to verify this potential. The learning gains identified in the results section can be targeted and measured in future research when empirically validating the use of immersive technologies for construction education. The contribution of this work is in synthesizing the learning gains that future researchers should target based on evidence from prior research in related learning contexts.more » « less
-
In a neuron network, synapses update individually using local information, allowing for entirely decentralized learning. In contrast, elements in an artificial neural network are typically updated simultaneously using a central processor. Here, we investigate the feasibility and effect of desynchronous learning in a recently introduced decentralized, physics-driven learning network. We show that desynchronizing the learning process does not degrade the performance for a variety of tasks in an idealized simulation. In experiment, desynchronization actually improves the performance by allowing the system to better explore the discretized state space of solutions. We draw an analogy between desynchronization and mini-batching in stochastic gradient descent and show that they have similar effects on the learning process. Desynchronizing the learning process establishes physics-driven learning networks as truly fully distributed learning machines, promoting better performance and scalability in deployment.more » « less
