# Transfer Learning by Modeling a Distribution over Policies

## TL;DR

This paper proposes a Bayesian deep reinforcement learning approach that models a distribution over policies to enhance exploration and transfer learning, demonstrating improved performance on GridWorld and MiniGrid environments.

## Contribution

It introduces a novel transfer strategy based on modeling a policy distribution, which promotes diversity and faster exploration in reinforcement learning.

## Key findings

- Faster exploration in transfer learning tasks.
- Improved performance on GridWorld and MiniGrid environments.
- Supports hypothesis that policy diversity aids transfer learning.

## Abstract

Exploration and adaptation to new tasks in a transfer learning setup is a central challenge in reinforcement learning. In this work, we build on the idea of modeling a distribution over policies in a Bayesian deep reinforcement learning setup to propose a transfer strategy. Recent works have shown to induce diversity in the learned policies by maximizing the entropy of a distribution of policies (Bachman et al., 2018; Garnelo et al., 2018) and thus, we postulate that our proposed approach leads to faster exploration resulting in improved transfer learning. We support our hypothesis by demonstrating favorable experimental results on a variety of settings on fully-observable GridWorld and partially observable MiniGrid (Chevalier-Boisvert et al., 2018) environments.