Online Limited Memory Neural-Linear Bandits with Likelihood Matching

Ofir Nabati, Tom Zahavy, Shie Mannor
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:7905-7915, 2021.

Abstract

We study neural-linear bandits for solving problems where {\em both} exploration and representation learning play an important role. Neural-linear bandits harnesses the representation power of Deep Neural Networks (DNNs) and combines it with efficient exploration mechanisms by leveraging uncertainty estimation of the model, designed for linear contextual bandits on top of the last hidden layer. In order to mitigate the problem of representation change during the process, new uncertainty estimations are computed using stored data from an unlimited buffer. Nevertheless, when the amount of stored data is limited, a phenomenon called catastrophic forgetting emerges. To alleviate this, we propose a likelihood matching algorithm that is resilient to catastrophic forgetting and is completely online. We applied our algorithm, Limited Memory Neural-Linear with Likelihood Matching (NeuralLinear-LiM2) on a variety of datasets and observed that our algorithm achieves comparable performance to the unlimited memory approach while exhibits resilience to catastrophic forgetting.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-nabati21a, title = {Online Limited Memory Neural-Linear Bandits with Likelihood Matching}, author = {Nabati, Ofir and Zahavy, Tom and Mannor, Shie}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {7905--7915}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/nabati21a/nabati21a.pdf}, url = {https://proceedings.mlr.press/v139/nabati21a.html}, abstract = {We study neural-linear bandits for solving problems where {\em both} exploration and representation learning play an important role. Neural-linear bandits harnesses the representation power of Deep Neural Networks (DNNs) and combines it with efficient exploration mechanisms by leveraging uncertainty estimation of the model, designed for linear contextual bandits on top of the last hidden layer. In order to mitigate the problem of representation change during the process, new uncertainty estimations are computed using stored data from an unlimited buffer. Nevertheless, when the amount of stored data is limited, a phenomenon called catastrophic forgetting emerges. To alleviate this, we propose a likelihood matching algorithm that is resilient to catastrophic forgetting and is completely online. We applied our algorithm, Limited Memory Neural-Linear with Likelihood Matching (NeuralLinear-LiM2) on a variety of datasets and observed that our algorithm achieves comparable performance to the unlimited memory approach while exhibits resilience to catastrophic forgetting.} }
Endnote
%0 Conference Paper %T Online Limited Memory Neural-Linear Bandits with Likelihood Matching %A Ofir Nabati %A Tom Zahavy %A Shie Mannor %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-nabati21a %I PMLR %P 7905--7915 %U https://proceedings.mlr.press/v139/nabati21a.html %V 139 %X We study neural-linear bandits for solving problems where {\em both} exploration and representation learning play an important role. Neural-linear bandits harnesses the representation power of Deep Neural Networks (DNNs) and combines it with efficient exploration mechanisms by leveraging uncertainty estimation of the model, designed for linear contextual bandits on top of the last hidden layer. In order to mitigate the problem of representation change during the process, new uncertainty estimations are computed using stored data from an unlimited buffer. Nevertheless, when the amount of stored data is limited, a phenomenon called catastrophic forgetting emerges. To alleviate this, we propose a likelihood matching algorithm that is resilient to catastrophic forgetting and is completely online. We applied our algorithm, Limited Memory Neural-Linear with Likelihood Matching (NeuralLinear-LiM2) on a variety of datasets and observed that our algorithm achieves comparable performance to the unlimited memory approach while exhibits resilience to catastrophic forgetting.
APA
Nabati, O., Zahavy, T. & Mannor, S.. (2021). Online Limited Memory Neural-Linear Bandits with Likelihood Matching. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:7905-7915 Available from https://proceedings.mlr.press/v139/nabati21a.html.

Related Material