decomposed mutual information optimization
Appendix of Decomposed Mutual Information Optimization for Generalized Context in Meta-Reinforcement Learning Y ao Mu The University of Hong Kong
Ping Luo is the corresponding author. With Equation 3 and Jensen's inequality applied in Equation 1, we have I (x,y) E Therefore, if the number of confounders increases, then the demand for data will grow exponentially. When data is not rich enough, the nesseray condition may not be satisfied. We provide the pseudo-code of DOMINO combined with model-based methods. Firstly, the past state-action pairs are encoded into the disentangled context vectors by the context encoder. Initialize batch B . for i = 1 to B do sample V Listing 1: PyTorch-style pseudo-code for dynamics change based on Mujoco engine.
Appendix of Decomposed Mutual Information Optimization for Generalized Context in Meta-Reinforcement Learning Y ao Mu The University of Hong Kong
Ping Luo is the corresponding author. With Equation 3 and Jensen's inequality applied in Equation 1, we have I (x,y) E Therefore, if the number of confounders increases, then the demand for data will grow exponentially. When data is not rich enough, the nesseray condition may not be satisfied. We provide the pseudo-code of DOMINO combined with model-based methods. Firstly, the past state-action pairs are encoded into the disentangled context vectors by the context encoder. Initialize batch B . for i = 1 to B do sample V Listing 1: PyTorch-style pseudo-code for dynamics change based on Mujoco engine.
DOMINO: Decomposed Mutual Information Optimization for Generalized Context in Meta-Reinforcement Learning
Adapting to the changes in transition dynamics is essential in robotic applications. By learning a conditional policy with a compact context, context-aware meta-reinforcement learning provides a flexible way to adjust behavior according to dynamics changes. However, in real-world applications, the agent may encounter complex dynamics changes. Multiple confounders can influence the transition dynamics, making it challenging to infer accurate context for decision-making. This paper addresses such a challenge by decomposed mutual information optimization (DOMINO) for context learning, which explicitly learns a disentangled context to maximize the mutual information between the context and historical trajectories while minimizing the state transition prediction error. Our theoretical analysis shows that DOMINO can overcome the underestimation of the mutual information caused by multi-confounded challenges via learning disentangled context and reduce the demand for the number of samples collected in various environments.