Article (Scientific journals)
On The Transferability of Deep-Q Networks
Sabatelli, Matthia; Geurts, Pierre
2021In Deep Reinforcement Learning Workshop of the 35th Conference on Neural Information Processing Systems
Peer reviewed
 

Files


Full Text
drlw_neurips_2021.pdf
Publisher postprint (728.65 kB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Deep Reinforcement Learning; Transfer Learning; Model-free Deep Reinforcement Learning
Abstract :
[en] Transfer Learning (TL) is an efficient machine learning paradigm that allows overcoming some of the hurdles that characterize the successful training of deep neural networks, ranging from long training times to the needs of large datasets. While exploiting TL is a well established and successful training practice in Supervised Learning (SL), its applicability in Deep Reinforcement Learning (DRL) is rarer. In this paper, we study the level of transferability of three different variants of Deep-Q Networks on popular DRL benchmarks as well as on a set of novel, carefully designed control tasks. Our results show that transferring neural networks in a DRL context can be particularly challenging and is a process which in most cases results in negative transfer. In the attempt of understanding why Deep-Q Networks transfer so poorly, we gain novel insights into the training dynamics that characterizes this family of algorithms.
Disciplines :
Computer science
Author, co-author :
Sabatelli, Matthia ;  Université de Liège - ULiège > Montefiore Institute
Geurts, Pierre ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Algorith. des syst. en interaction avec le monde physique
Language :
English
Title :
On The Transferability of Deep-Q Networks
Publication date :
December 2021
Journal title :
Deep Reinforcement Learning Workshop of the 35th Conference on Neural Information Processing Systems
Peer reviewed :
Peer reviewed
Available on ORBi :
since 27 December 2021

Statistics


Number of views
66 (3 by ULiège)
Number of downloads
22 (2 by ULiège)

Bibliography


Similar publications



Contact ORBi