# Actively Learning Gaussian Process Dynamics

arXiv: https://arxiv.org/abs/1911.09946 · HTML: https://mlsanity.com/paper/1911.09946/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Nov 2019.

## Built on by (most cited first, 26 of 26)

- [Optimistic Active Exploration of Dynamical Systems](https://mlsanity.com/paper/2306.12371/) (2023)
- [An Experimental Design Perspective on Model-Based Reinforcement Learning](https://mlsanity.com/paper/2112.05244/) (2021)
- [Structure-preserving Gaussian Process Dynamics](https://mlsanity.com/paper/2102.01606/) (2021)
- [SOMBRL: Scalable and Optimistic Model-Based RL](https://mlsanity.com/paper/2511.20066/) (2025)
- [Explore the Context: Optimal Data Collection for Context-Conditional Dynamics Models](https://mlsanity.com/paper/2102.11394/) (2021)
- [Probabilistic Robust Linear Quadratic Regulators with Gaussian Processes](https://mlsanity.com/paper/2105.07668/) (2021)
- [On the Consistency of Kernel Methods with Dependent Observations](https://mlsanity.com/paper/2406.06101/) (2024)
- [On Foundation Models for Dynamical Systems from Purely Synthetic Data](https://mlsanity.com/paper/2412.00395/) (2024)
- [Out of Distribution Detection via Domain-Informed Gaussian Process State Space Models](https://mlsanity.com/paper/2309.06655/) (2023)
- [Active Symbolic Discovery of Ordinary Differential Equations via Phase Portrait Sketching](https://mlsanity.com/paper/2409.01416/) (2024)
- [First, Learn What You Don't Know: Active Information Gathering for Driving at the Limits of Handling](https://mlsanity.com/paper/2411.00107/) (2024)
- [Active Learning in Gaussian Process State Space Model](https://mlsanity.com/paper/2108.00819/) (2021)
- [The Past Does Matter: Correlation of Subsequent States in Trajectory Predictions of Gaussian Process Models](https://mlsanity.com/paper/2211.11103/) (2022)
- [FLEX: an Adaptive Exploration Algorithm for Nonlinear Systems](https://mlsanity.com/paper/2304.13426/) (2023)
- [Assessment and treatment of visuospatial neglect using active learning with Gaussian processes regression](https://mlsanity.com/paper/2310.13701/) (2023)
- [Learning Soft Robotic Dynamics with Active Exploration](https://mlsanity.com/paper/2510.27428/) (2025)
- [Deep active learning for nonlinear system identification](https://mlsanity.com/paper/2302.12667/) (2023)
- [Efficient Sensor Placement from Regression with Sparse Gaussian Processes in Continuous and Discrete Spaces](https://mlsanity.com/paper/2303.00028/) (2023)
- [Future Aware Safe Active Learning of Time Varying Systems using Gaussian Processes](https://mlsanity.com/paper/2405.10581/) (2024)
- [Active Sensing and Deferred-Decision Trajectory Optimization for Robust Target Identification](https://mlsanity.com/paper/2606.22277/) (2026)
- [Learning Disturbances Online for Risk-Aware Control: Risk-Aware Flight with Less Than One Minute of Data](https://mlsanity.com/paper/2212.06253/) (2022)
- [Episodic Gaussian Process-Based Learning Control with Vanishing Tracking Errors](https://mlsanity.com/paper/2307.04415/) (2023)
- [Comparison of Data Reduction Criteria for Online Gaussian Processes](https://mlsanity.com/paper/2508.10815/) (2025)
- [Worst-Case Distance-Aware Error Bounds for Neural Networks](https://mlsanity.com/paper/2510.22021/) (2025)
- [Generalized Information Gathering Under Dynamics Uncertainty](https://mlsanity.com/paper/2601.21988/) (2026)
- [All Models are Wrong, Knowing Where is Useful: On Model Uncertainty in Reinforcement Learning](https://mlsanity.com/paper/2606.01363/) (2026)

## Abstract (arXiv)

Despite the availability of ever more data enabled through modern sensor and computer technology, it still remains an open problem to learn dynamical systems in a sample-efficient way. We propose active learning strategies that leverage information-theoretical properties arising naturally during Gaussian process regression, while respecting constraints on the sampling process imposed by the system dynamics. Sample points are selected in regions with high uncertainty, leading to exploratory behavior and data-efficient training of the model. All results are finally verified in an extensive numerical benchmark.