TY - JOUR

T1 - Improved regret bounds for bandit combinatorial optimization

AU - Ito, Shinji

AU - Hatano, Daisuke

AU - Sumita, Hanna

AU - Takemura, Kei

AU - Fukunaga, Takuro

AU - Kakimura, Naonori

AU - Kawarabayashi, Ken Ichi

N1 - Funding Information:
∗This work was supported by JST, ERATO, Grant Number JPMJER1201, Japan. †This work was supported by JST, ACT-I, Grant Number JPMJPR18U5, Japan. ‡This work was supported by JST, PRESTO, Grant Number JPMJPR1759, Japan. §This work was supported by JSPS, KAKENHI, Grant Number JP18H05291, Japan.

PY - 2019

Y1 - 2019

N2 - Bandit combinatorial optimization is a bandit framework in which a player chooses an action within a given finite set A ? {0, 1}d and incurs a loss that is the inner product of the chosen action and an unobservable loss vector in Rd in each round. In this paper, we aim to reveal the property, which makes the bandit combinatorial optimization hard. Recently, Cohen et al. [8] obtained a lower bound ?(pdk3T/log T) of the regret, where k is the maximum `1-norm of action vectors, and T is the number of rounds. This lower bound was achieved by considering a continuous strongly-correlated distribution of losses. Our main contribution is that we managed to improve this bound by ?(vdk3T) through applying a factor of vlog T, which can be done by means of strongly-correlated losses with binary values. The bound derives better regret bounds for three specific examples of the bandit combinatorial optimization: the multitask bandit, the bandit ranking and the multiple-play bandit. In particular, the bound obtained for the bandit ranking in the present study addresses an open problem raised in [8]. In addition, we demonstrate that the problem becomes easier without considering correlations among entries of loss vectors. In fact, if each entry of loss vectors is an independent random variable, then, one can achieve a regret of Õ(vdk2T), which is vk times smaller than the lower bound shown above. The observed results indicated that correlation among losses is the reason for observing a large regret.

AB - Bandit combinatorial optimization is a bandit framework in which a player chooses an action within a given finite set A ? {0, 1}d and incurs a loss that is the inner product of the chosen action and an unobservable loss vector in Rd in each round. In this paper, we aim to reveal the property, which makes the bandit combinatorial optimization hard. Recently, Cohen et al. [8] obtained a lower bound ?(pdk3T/log T) of the regret, where k is the maximum `1-norm of action vectors, and T is the number of rounds. This lower bound was achieved by considering a continuous strongly-correlated distribution of losses. Our main contribution is that we managed to improve this bound by ?(vdk3T) through applying a factor of vlog T, which can be done by means of strongly-correlated losses with binary values. The bound derives better regret bounds for three specific examples of the bandit combinatorial optimization: the multitask bandit, the bandit ranking and the multiple-play bandit. In particular, the bound obtained for the bandit ranking in the present study addresses an open problem raised in [8]. In addition, we demonstrate that the problem becomes easier without considering correlations among entries of loss vectors. In fact, if each entry of loss vectors is an independent random variable, then, one can achieve a regret of Õ(vdk2T), which is vk times smaller than the lower bound shown above. The observed results indicated that correlation among losses is the reason for observing a large regret.

UR - http://www.scopus.com/inward/record.url?scp=85090170188&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85090170188&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85090170188

VL - 32

JO - Advances in Neural Information Processing Systems

JF - Advances in Neural Information Processing Systems

SN - 1049-5258

T2 - 33rd Annual Conference on Neural Information Processing Systems, NeurIPS 2019

Y2 - 8 December 2019 through 14 December 2019

ER -