Semantic Scholar Open Access 2018 175 sitasi

Efficient Exploration Through Bayesian Deep Q-Networks

K. Azizzadenesheli E. Brunskill Anima Anandkumar

Lihat Sumber DOI

Abstrak

We propose Bayesian Deep Q-Network (BDQN), a practical Thompson sampling based Reinforcement Learning (RL) Algorithm. Thompson sampling allows for targeted exploration in high dimensions through posterior sampling but is usually computationally expensive. We address this limitation by introducing uncertainty only at the output layer of the network through a Bayesian Linear Regression (BLR) model, which can be trained with fast closed-form updates and its samples can be drawn efficiently through the Gaussian distribution. We apply our method to a wide range of Atari games in Arcade Learning Environments. Since BDQN carries out more efficient exploration, it is able to reach higher rewards substantially faster than a key baseline, double deep Q network DDQN.

Topik & Kata Kunci

Computer Science Mathematics

Penulis (3)

K. Azizzadenesheli

E. Brunskill

Anima Anandkumar

Format Sitasi

APA MLA BibTeX

Azizzadenesheli, K., Brunskill, E., Anandkumar, A. (2018). Efficient Exploration Through Bayesian Deep Q-Networks. https://doi.org/10.1109/ITA.2018.8503252

Akses Cepat

Lihat di Sumber doi.org/10.1109/ITA.2018.8503252

Informasi Jurnal

Tahun Terbit: 2018
Bahasa: en
Total Sitasi: 175×
Sumber Database: Semantic Scholar
DOI: 10.1109/ITA.2018.8503252
Akses: Open Access ✓