Composite Gaussian Processes Flows for Learning Discontinuous Multimodal Policies

2025-02-04

AI Summary

Addressing the challenge of jointly accommodating multimodality, local discontinuities, and computational efficiency in robot control policy learning, this paper proposes Composite Gaussian Process Flows (CGP-Flows)—the first semi-parametric policy modeling framework that synergistically integrates Overlapping Mixture of Gaussian Processes (OMGPs) with Continuous Normalizing Flows (CNFs). CGP-Flows preserves the efficient Bayesian inference and structural interpretability of OMGPs while leveraging CNFs’ strong representational capacity for nonlinear and discontinuous mappings. Experiments on both simulated and real-robot tasks demonstrate that CGP-Flows significantly improves policy success rates. Statistical validation via χ² tests confirms its superiority over all baselines at *p* < 0.01. The method achieves state-of-the-art performance across diverse benchmarks, offering a principled trade-off between expressivity, uncertainty quantification, and scalability in robotic policy learning.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Machine Learning: Adversarial Learning & Robustness
• Planning, Routing, and Scheduling: Mixed Discrete/Continuous Planning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Learning control policies for real-world robotic tasks often involve challenges such as multimodality, local discontinuities, and the need for computational efficiency. These challenges arise from the complexity of robotic environments, where multiple solutions may coexist. To address these issues, we propose Composite Gaussian Processes Flows (CGP-Flows), a novel semi-parametric model for robotic policy. CGP-Flows integrate Overlapping Mixtures of Gaussian Processes (OMGPs) with the Continuous Normalizing Flows (CNFs), enabling them to model complex policies addressing multimodality and local discontinuities. This hybrid approach retains the computational efficiency of OMGPs while incorporating the flexibility of CNFs. Experiments conducted in both simulated and real-world robotic tasks demonstrate that CGP-flows significantly improve performance in modeling control policies. In a simulation task, we confirmed that CGP-Flows had a higher success rate compared to the baseline method, and the success rate of GCP-Flow was significantly different from the success rate of other baselines in chi-square tests.

Problem

• Learning discontinuous multimodal control policies
• Addressing computational efficiency in robotics
• Improving performance in robotic tasks

Innovation

• Composite Gaussian Processes Flows
• Overlapping Mixtures of Gaussian Processes
• Continuous Normalizing Flows