Event-triggered learning
From MaRDI portal
Abstract: The efficient exchange of information is an essential aspect of intelligent collective behavior. Event-triggered control and estimation achieve some efficiency by replacing continuous data exchange between agents with intermittent, or event-triggered communication. Typically, model-based predictions are used at times of no data transmission, and updates are sent only when the prediction error grows too large. The effectiveness in reducing communication thus strongly depends on the quality of the prediction model. In this article, we propose event-triggered learning as a novel concept to reduce communication even further and to also adapt to changing dynamics. By monitoring the actual communication rate and comparing it to the one that is induced by the model, we detect a mismatch between model and reality and trigger model learning when needed. Specifically, for linear Gaussian dynamics, we derive different classes of learning triggers solely based on a statistical analysis of inter-communication times and formally prove their effectiveness with the aid of concentration inequalities.
Recommendations
- Adaptive event-triggered distributed model predictive control for multi-agent systems
- Event-triggered feedback in control, estimation, and optimization
- An event-triggered control approach for the leader-tracking problem with heterogeneous agents
- Event-triggered distributed estimation with decaying communication rate
- Distributed event-triggered unadjusted Langevin algorithm for Bayesian learning
Cites work
- scientific article; zbMATH DE number 54145 (Why is no real title available?)
- A distributed Kalman filter with event-triggered communication and guaranteed stability
- Anticipatory behavior in adaptive learning systems. Foundations, theories, and systems.
- Energy-based event-triggered state estimation for hidden Markov models
- Event Based State Estimation With Time Synchronous Updates
- Event-Based Sensor Data Scheduling: Trade-Off Between Communication Rate and Estimation Quality
- Event-Based State Estimation With Variance-Based Triggering
- Event-triggered feedback in control, estimation, and optimization
- Event-triggered maximum likelihood state estimation
- First exit time probability for multidimensional diffusions: A PDE-based approach
- Model-free event-triggered control algorithm for continuous-time linear systems with optimal performance
- Probability Inequalities for Sums of Bounded Random Variables
- Process fault detection based on modeling and estimation methods - a survey
- Robust event-triggered output feedback learning algorithm for voltage source inverters with unknown load and parameter variations
- Stochastic Event-Triggered Sensor Schedule for Remote State Estimation
- Stochastic differential equations. An introduction with applications.
- Stochastic processes and applications. Diffusion processes, the Fokker-Planck and Langevin equations
- The significance probability of the Smirnov two-sample test
- The tight constant in the Dvoretzky-Kiefer-Wolfowitz inequality
Cited in
(6)- Robust MPC with event-triggered learning for unknown linear time-varying systems
- Transactional events for ML
- Robust event-driven interactions in cooperative multi-agent learning
- Model‐free self‐triggered control based on deep reinforcement learning for unknown nonlinear systems
- Current trends and future directions in event-based control
- Timely Event Detection by Networked Learners
This page was built for publication: Event-triggered learning
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2184557)
