Causal Navigation by Continuous-time Neural Networks

Abstract

Imitation learning enables high-fidelity, vision-based learning of policies within rich, photorealistic environments. However, such techniques often rely on traditional discrete-time neural models and face difficulties in generalizing to domain shifts by failing to account for the causal relationships between the agent and the environment. In this paper, we propose a theoretical and experimental framework for learning causal representations using continuous-time neural networks, specifically over their discrete-time counterparts. We evaluate our method in the context of visual-control learning of drones over a series of complex tasks, ranging from short- and long-term navigation, to chasing static and dynamic objects through photorealistic environments. Our results demonstrate that causal continuous-time deep models can perform robust navigation tasks, where advanced recurrent models fail. These models learn complex causal control representations directly from raw visual inputs and scale to solve a variety of tasks using imitation learning.

24 Pages

References in corpus (11)

• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Learning to Explore using Active Neural SLAM
• Continuous Time Bayesian Networks
• Latent ODEs for Irregularly-Sampled Time Series
• ANODE: Unconditionally Accurate Memory-Efficient Gradients for Neural ODEs
• Unifying Map and Landmark Based Representations for Visual Navigation
• Can Autonomous Vehicles Identify, Recover From, and Adapt to Distribution Shifts?
• Learning to Control PDEs with Differentiable Physics
• From Ordinary Differential Equations to Structural Causal Models: the deterministic case
• Discrete Event, Continuous Time RNNs
• Adversarial Imitation Learning from Incomplete Demonstrations