Equivalence Between Wasserstein and Value-Aware Loss for Model-based Reinforcement Learning

Abstract

Learning a generative model is a key component of model-based reinforcement learning. Though learning a good model in the tabular setting is a simple task, learning a useful model in the approximate setting is challenging. In this context, an important question is the loss function used for model learning as varying the loss function can have a remarkable impact on effectiveness of planning. Recently Farahmand et al. (2017) proposed a value-aware model learning (VAML) objective that captures the structure of value function during model learning. Using tools from Asadi et al. (2018), we show that minimizing the VAML objective is in fact equivalent to minimizing the Wasserstein metric. This equivalence improves our understanding of value-aware models, and also creates a theoretical foundation for applications of Wasserstein in model-based reinforcement~learning.

arXiv commentAccepted at the FAIM workshop "Prediction and Generative Modeling in Reinforcement Learning", Stockholm, Sweden, 2018

References in corpus (4)

• Learning with a Wasserstein Loss
• Bridging the Gap Between Value and Policy Based Reinforcement Learning
• A unified view of entropy-regularized Markov decision processes
• Lipschitz Continuity in Model-based Reinforcement Learning

Cited by in corpus (6)

• Formal verification of higher-order probabilistic programs
• Combating the Compounding-Error Problem with a Multi-step Model
• Gradient-Aware Model-based Policy Search
• The Value Equivalence Principle for Model-Based Reinforcement Learning
• Policy-Aware Model Learning for Policy Gradient Methods
• Towards a Simple Approach to Multi-step Model-based Reinforcement Learning