no code implementations • 26 Apr 2024 • Ken Yokoyama, Shinji Ito, Tatsuya Matsuoka, Kei Kimura, Makoto Yokoo
An existing general framework for dealing with such objective functions is the online submodular minimization.
no code implementations • 8 Mar 2024 • Jongyeong Lee, Junya Honda, Shinji Ito, Min-hwan Oh
In this paper, we establish a sufficient condition for perturbations to achieve $\mathcal{O}(\sqrt{KT})$ regrets in the adversarial setting, which covers, e. g., Fr\'{e}chet, Pareto, and Student-$t$ distributions.
no code implementations • 5 Mar 2024 • Masahiro Kato, Shinji Ito
For this issue, this study proposes an algorithm whose regret satisfies $O(\log(T))$ in the setting when the suboptimality gap is lower-bounded.
no code implementations • 1 Mar 2024 • Shinji Ito, Taira Tsuchiya, Junya Honda
Follow-The-Regularized-Leader (FTRL) is known as an effective and versatile approach in online learning, where appropriate choice of the learning rate is crucial for smaller regret.
no code implementations • 20 Feb 2024 • Taira Tsuchiya, Shinji Ito
We first prove that if an optimal decision is on the boundary of a feasible set and the gradient of an underlying loss function is non-zero, then the algorithm achieves a regret upper bound of $O(\rho \log T)$ in stochastic environments.
no code implementations • 13 Feb 2024 • Taira Tsuchiya, Shinji Ito, Junya Honda
This development allows us to significantly improve the existing regret bounds of best-of-both-worlds (BOBW) algorithms, which achieves nearly optimal bounds both in stochastic and adversarial environments.
no code implementations • 12 Feb 2024 • Junpei Komiyama, Shinji Ito, Yuichi Yoshida, Souta Koshino
For the analysis of these algorithms, we propose a principled approach to limiting the probability of nonreplication.
no code implementations • 27 Dec 2023 • Masahiro Kato, Shinji Ito
The goal of this study is to develop a strategy that is effective in both stochastic and adversarial environments, with theoretical guarantees.
no code implementations • 19 Dec 2023 • Koji Ichikawa, Shinji Ito, Daisuke Hatano, Hanna Sumita, Takuro Fukunaga, Naonori Kakimura, Ken-ichi Kawarabayashi
Firstly, we show that a mixture distribution that has a greedy-applicable component is also greedy-applicable.
no code implementations • NeurIPS 2023 • Taira Tsuchiya, Shinji Ito, Junya Honda
With this result, we establish several algorithms with three types of adaptivity: sparsity, game-dependency, and best-of-both-worlds (BOBW).
no code implementations • 24 Feb 2023 • Shinji Ito, Kei Takemura
At the higher level, the proposed algorithm adapts to a variety of types of environments.
no code implementations • 29 Jul 2022 • Taira Tsuchiya, Shinji Ito, Junya Honda
This study considers the partial monitoring problem with $k$-actions and $d$-outcomes and provides the first best-of-both-worlds algorithms, whose regrets are favorably bounded both in the stochastic and adversarial regimes.
no code implementations • 14 Jun 2022 • Shinji Ito, Taira Tsuchiya, Junya Honda
In fact, they have provided a stochastic MAB algorithm with gap-variance-dependent regret bounds of $O(\sum_{i: \Delta_i>0} (\frac{\sigma_i^2}{\Delta_i} + 1) \log T )$ for loss variance $\sigma_i^2$ of arm $i$.
no code implementations • 2 Jun 2022 • Shinji Ito, Taira Tsuchiya, Junya Honda
As Alon et al. [2015] have shown, tight regret bounds depend on the structure of the feedback graph: strongly observable graphs yield minimax regret of $\tilde{\Theta}( \alpha^{1/2} T^{1/2} )$, while weakly observable graphs induce minimax regret of $\tilde{\Theta}( \delta^{1/3} T^{2/3} )$, where $\alpha$ and $\delta$, respectively, represent the independence number of the graph and the domination number of a certain portion of the graph.
no code implementations • 15 Mar 2022 • Hanna Sumita, Shinji Ito, Kei Takemura, Daisuke Hatano, Takuro Fukunaga, Naonori Kakimura, Ken-ichi Kawarabayashi
The key features of our problem are (1) an agent is reusable, i. e., an agent comes back to the market after completing the assigned task, (2) an agent may reject the assigned task to stay the market, and (3) a task may accommodate multiple agents.
no code implementations • NeurIPS 2021 • Shinji Ito
This study aims to develop bandit algorithms that automatically exploit tendencies of certain environments to improve performance, without any prior knowledge regarding the environments.
no code implementations • NeurIPS 2021 • Shinji Ito
The main contribution of this paper is to show that optimal robustness can be expressed by a square-root dependency on the amount of corruption.
no code implementations • 20 Jan 2021 • Kei Takemura, Shinji Ito, Daisuke Hatano, Hanna Sumita, Takuro Fukunaga, Naonori Kakimura, Ken-ichi Kawarabayashi
However, there is a gap of $\tilde{O}(\max(\sqrt{d}, \sqrt{k}))$ between the current best upper and lower bounds, where $d$ is the dimension of the feature vectors, $k$ is the number of the chosen arms in a round, and $\tilde{O}(\cdot)$ ignores the logarithmic factors.
no code implementations • NeurIPS 2020 • Shinji Ito, Daisuke Hatano, Hanna Sumita, Kei Takemura, Takuro Fukunaga, Naonori Kakimura, Ken-ichi Kawarabayashi
This paper offers a nearly optimal algorithm for online linear optimization with delayed bandit feedback.
no code implementations • NeurIPS 2020 • Shinji Ito
Swap regret, a generic performance measure of online decision-making algorithms, plays an important role in the theory of repeated games, along with a close connection to correlated equilibria in strategic games.
no code implementations • NeurIPS 2020 • Shinji Ito, Shuichi Hirahara, Tasuku Soma, Yuichi Yoshida
We propose novel algorithms with first- and second-order regret bounds for adversarial linear bandits.
no code implementations • NeurIPS 2019 • Shinji Ito
This paper considers submodular function minimization with \textit{noisy evaluation oracles} that return the function value of a submodular objective with zero-mean additive noise.
no code implementations • NeurIPS 2019 • Shinji Ito, Daisuke Hatano, Hanna Sumita, Kei Takemura, Takuro Fukunaga, Naonori Kakimura, Ken-ichi Kawarabayashi
Our algorithm for non-stochastic settings has an oracle complexity of $\tilde{O}( T )$ and is the first algorithm that achieves both a regret bound of $\tilde{O}( \sqrt{T} )$ and an oracle complexity of $\tilde{O} ( \mathrm{poly} ( T ) )$, given only linear optimization oracles.
no code implementations • NeurIPS 2019 • Shinji Ito, Daisuke Hatano, Hanna Sumita, Kei Takemura, Takuro Fukunaga, Naonori Kakimura, Ken-ichi Kawarabayashi
\textit{Bandit combinatorial optimization} is a bandit framework in which a player chooses an action within a given finite set $\mathcal{A} \subseteq \{ 0, 1 \}^d$ and incurs a loss that is the inner product of the chosen action and an unobservable loss vector in $\mathbb{R} ^ d$ in each round.
no code implementations • 5 Sep 2019 • Kei Takemura, Shinji Ito
Our empirical evaluation with artificial and real-world datasets demonstrates that the proposed algorithms with the arm-wise randomization technique outperform the existing algorithms without this technique, especially for the clustered case.
no code implementations • NeurIPS 2018 • Shinji Ito, Daisuke Hatano, Sumita Hanna, Akihiro Yabe, Takuro Fukunaga, Naonori Kakimura, Ken-ichi Kawarabayashi
Online portfolio selection is a sequential decision-making problem in which a learner repetitively selects a portfolio over a set of assets, aiming to maximize long-term return.
no code implementations • ICML 2018 • Shinji Ito, Akihiro Yabe, Ryohei Fujimaki
Predictive optimization, however, suffers from the problem of a calculated optimal solution’s being evaluated too optimistically, i. e., the value of the objective function is overestimated.
no code implementations • ICML 2018 • Akihiro Yabe, Daisuke Hatano, Hanna Sumita, Shinji Ito, Naonori Kakimura, Takuro Fukunaga, Ken-ichi Kawarabayashi
In this setting, the arms are identified with interventions on a given causal graph, and the effect of an intervention propagates throughout all over the causal graph.
no code implementations • NeurIPS 2017 • Shinji Ito, Daisuke Hatano, Hanna Sumita, Akihiro Yabe, Takuro Fukunaga, Naonori Kakimura, Ken-ichi Kawarabayashi
Under these assumptions, we present polynomial-time sublinear-regret algorithms for the online sparse linear regression.
no code implementations • NeurIPS 2016 • Shinji Ito, Ryohei Fujimaki
On the basis of this connection, we propose an efficient algorithm that employs network flow algorithms.
no code implementations • 18 May 2016 • Shinji Ito, Ryohei Fujimaki
This paper addresses a novel data science problem, prescriptive price optimization, which derives the optimal price strategy to maximize future profit/revenue on the basis of massive predictive formulas produced by machine learning.