We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Exploring Variational Deep Q Networks

Abstract: This study provides both analysis and a refined, research-ready implementation of Tang and Kucukelbir's Variational Deep Q Network, a novel approach to maximising the efficiency of exploration in complex learning environments using Variational Bayesian Inference. Alongside reference implementations of both Traditional and Double Deep Q Networks, a small novel contribution is presented - the Double Variational Deep Q Network, which incorporates improvements to increase the stability and robustness of inference-based learning. Finally, an evaluation and discussion of the effectiveness of these approaches is discussed in the wider context of Bayesian Deep Learning.
Comments: 12 pages, 5 figures
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Cite as: arXiv:2008.01641 [cs.LG]
  (or arXiv:2008.01641v1 [cs.LG] for this version)

Submission history

From: Alexander Harri Bell-Thomas [view email]
[v1] Tue, 4 Aug 2020 15:36:31 GMT (10063kb,D)

Link back to: arXiv, form interface, contact.