Imperial College London
Alternate

Professor Anil Anthony Bharath

Faculty of Engineering, Department of Bioengineering

Academic Director (Singapore)
 
 
 
//

Contact

 

+44 (0)20 7594 5463a.bharath Website

 
 
//

Location

 

4.12Royal School of MinesSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@unpublished{Agostinelli:2019,
author = {Agostinelli, A and Arulkumaran, K and Sarrico, M and Richemond, P and Bharath, AA},
title = {Memory-efficient episodic control reinforcement learning with dynamic online k-means},
url = {http://arxiv.org/abs/1911.09560v1},
year = {2019}
}
Download

RIS format (EndNote, RefMan)

TY  - UNPB
AB  - Recently, neuro-inspired episodic control (EC) methods have been developed toovercome the data-inefficiency of standard deep reinforcement learningapproaches. Using non-/semi-parametric models to estimate the value function,they learn rapidly, retrieving cached values from similar past states. Inrealistic scenarios, with limited resources and noisy data, maintainingmeaningful representations in memory is essential to speed up the learning andavoid catastrophic forgetting. Unfortunately, EC methods have a large space andtime complexity. We investigate different solutions to these problems based onprioritising and ranking stored states, as well as online clusteringtechniques. We also propose a new dynamic online k-means algorithm that is bothcomputationally-efficient and yields significantly better performance atsmaller memory sizes; we validate this approach on classic reinforcementlearning environments and Atari games.
AU  - Agostinelli,A
AU  - Arulkumaran,K
AU  - Sarrico,M
AU  - Richemond,P
AU  - Bharath,AA
PY  - 2019///
TI  - Memory-efficient episodic control reinforcement learning with dynamic online k-means
UR  - http://arxiv.org/abs/1911.09560v1
UR  - https://arxiv.org/abs/1911.09560v1
UR  - http://hdl.handle.net/10044/1/75281
ER  -
Download