VUSFA:Variational Universal Successor Features Approximator to Improve Transfer DRL for Target Driven Visual Navigation

Shamane Siriwardhana, Rivindu Weerasakera, Denys J. C. Matthies, Suranga Nanayakkara
8/18/2019
cs.LGcs.AIcs.NE

Official code: shamanez/VUSFA-Variational-Universal-Successor-Features-Approximator (14 stars, last commit 2021-03-31).

Abstract

In this paper, we show how novel transfer reinforcement learning techniques can be applied to the complex task of target driven navigation using the photorealistic AI2THOR simulator. Specifically, we build on the concept of Universal Successor Features with an A3C agent. We introduce the novel architectural contribution of a Successor Feature Dependant Policy (SFDP) and adopt the concept of Variational Information Bottlenecks to achieve state of the art performance. VUSFA, our final architecture, is a straightforward approach that can be implemented using our open source repository. Our approach is generalizable, showed greater stability in training, and outperformed recent approaches in terms of transfer learning ability.

View on arXivView PDF

Code Implementations(1)

shamanez/VUSFA-Variational-Universal-Successor-Features-ApproximatorOfficial100%

This repository contains implementations of the paper VUSFA

142PythonFeb 25, 20195 years ago

Similar Papers

Value Explicit Pretraining for Learning Transferable Representations

85% match2023

Trajectory-Diversity-Driven Robust Vision-and-Language Navigation

84% match1 repo42026

VR-Robo: A Real-to-Sim-to-Real Framework for Visual Robot Navigation and Locomotion

83% match1 repo1922025

Task and Domain Adaptive Reinforcement Learning for Robot Control

82% match1 repo42024

Transfer Learning for Related Reinforcement Learning Tasks via Image-to-Image Translation

82% match1 repo512018

PoliFormer: Scaling On-Policy RL with Transformers Results in Masterful Navigators

82% match1 repo1062024

Cite this paper

@article{siriwardhana2019vusfavariational,
  title  = {VUSFA:Variational Universal Successor Features Approximator to Improve Transfer DRL for Target Driven Visual Navigation},
  author = {Shamane Siriwardhana and Rivindu Weerasakera and Denys J. C. Matthies and Suranga Nanayakkara},
  year   = {2019},
  eprint = {1908.06376},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1908.06376}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1908.06376.svg)](https://codeofpaper.com/paper/1908.06376)

Discussion