Distributed Multitask Reinforcement Learning with Quadratic Convergence

back to our research

Distributed Multitask Reinforcement Learning with Quadratic Convergence

December 2 - 8, 2018 NeurIPS, Montreal, Canada

Authors: Rasul Tutunov (PROWLER.io), Dongho Kim (PROWLER.io), Haitham Bou-Ammar (PROWLER.io)

Abstract: Multitask reinforcement learning (MTRL) suffers from scalability issues when the number of tasks or trajectories grows large. The main reason behind this drawback is the reliance on centeralised solutions. Recent methods exploited the connection between MTRL and general consensus to propose scalable solutions. These methods, however, suffer from two drawbacks. First, they rely on predefined objectives, and, second, exhibit linear convergence guarantees. In this paper, we improve over state-of-the-art by deriving multitask reinforcement learning from a variational inference perspective. We then propose a novel distributed solver for MTRL with quadratic convergence guarantees.

Reinforcement Learning

NeurIPS

Multitask Learning

Distributed Optimisation

Data Efficiency

Scalability


See paper

Join us to make AI that will change the world

join our team