python

Learning representations for reinforcement learning (RL) has shown much promise for continuous control. In this project, we investigate using vector quantization to prevent representation collapse when learning representations for RL using a self-supervised latent-state consistency loss.

This work presents a learning-based control method for navigating to a target state in unknown, or partially unknown, multimodal dynamical systems. In particular, it develops a model-based reinforcement learning algorithm that can remain in a desired dynamics mode with high probability. For example, if some of the dynamics modes are believed to be inoperable.

Mode remaining navigation (and exploration) in unknown multimodal dynamical systems via model-based reinforcement learning.

In this talk I will present our ICRA 2021 paper “Trajectory Optimisation in Learned Multimodal Dynamical Systems via Latent-ODE Collocation”.

Mixture models are inherently unidentifiable as different combinations of component distributions and mixture weights can generate the …

Minimal Gaussian process library in JAX with a simple (custom) approach to state management.

Synergising Bayesian inference and Riemannian geometry for control in multimodal dynamical systems.

This talk presented recent work synergising Bayesian inference and probabilistic Riemannian geometries to control multimodal dynamical …

This work presents a two-stage method to perform trajectory optimisation in multimodal dynamical systems with unknown nonlinear stochastic transition dynamics. The method finds trajectories that remain in a preferred dynamics mode where possible and in regions of the transition dynamics model that have been observed and can be predicted confidently.

This work introduces a variational lower bound for the Mixture of Gaussian Process Experts model with a GP-based gating network based on sparse GPs. The model (and inference) are implemented in GPflow/TensorFlow.