Search Results for author: Eiji Takimoto

Found 12 papers, 3 papers with code

Online Combinatorial Linear Optimization via a Frank-Wolfe-based Metarounding Algorithm

no code implementations19 Oct 2023 Ryotaro Mitsuboshi, Kohei Hatano, Eiji Takimoto

Metarounding is an approach to convert an approximation algorithm for linear optimization over some combinatorial classes to an online linear optimization algorithm for the same class.

Boosting-based Construction of BDDs for Linear Threshold Functions and Its Application to Verification of Neural Networks

no code implementations8 Jun 2023 Yiping Tang, Kohei Hatano, Eiji Takimoto

Some previous work proposes to transform neural networks into equivalent Boolean expressions and apply verification techniques for characteristics of interest.

Boosting as Frank-Wolfe

1 code implementation22 Sep 2022 Ryotaro Mitsuboshi, Kohei Hatano, Eiji Takimoto

LPBoost rapidly converges to an $\epsilon$-approximate solution in practice, but it is known to take $\Omega(m)$ iterations in the worst case, where $m$ is the sample size.

A generalised log-determinant regularizer for online semi-definite programming and its applications

no code implementations10 Dec 2020 Yaxiong Liu, Ken-ichiro Moridomi, Kohei Hatano, Eiji Takimoto

We consider a variant of online semi-definite programming problem (OSDP): The decision space consists of semi-definite matrices with bounded $\Gamma$-trace norm, which is a generalization of trace norm defined by a positive definite matrix $\Gamma.$ To solve this problem, we utilise the follow-the-regularized-leader algorithm with a $\Gamma$-dependent log-determinant regularizer.

Matrix Completion

Improved algorithms for online load balancing

no code implementations15 Jul 2020 Yaxiong Liu, Kohei Hatano, Eiji Takimoto

The cost is the makespan if the norm is $L_\infty$-norm.

Theory and Algorithms for Shapelet-based Multiple-Instance Learning

1 code implementation31 May 2020 Daiki Suehiro, Kohei Hatano, Eiji Takimoto, Shuji Yamamoto, Kenichi Bannai, Akiko Takeda

We propose a new formulation of Multiple-Instance Learning (MIL), in which a unit of data consists of a set of instances called a bag.

Multiple Instance Learning Time Series Analysis +1

Simplified and Unified Analysis of Various Learning Problems by Reduction to Multiple-Instance Learning

1 code implementation14 Nov 2019 Daiki Suehiro, Eiji Takimoto

In this work, we focus on a particular problem formulation called Multiple-Instance Learning (MIL), and show that various learning problems including all the problems mentioned above with some of new problems can be reduced to MIL with theoretically guaranteed generalization bounds, where the reductions are established under a new reduction scheme we provide as a by-product.

Generalization Bounds Multi-Label Learning +2

Online linear optimization with the log-determinant regularizer

no code implementations27 Oct 2017 Ken-ichiro Moridomi, Kohei Hatano, Eiji Takimoto

Moreover, we apply our method to online linear optimization over vectors and show that the FTRL with the Burg entropy regularizer, which is the analogue of the log-determinant regularizer in the vector case, works well.

Collaborative Filtering

Bandit Online Optimization Over the Permutahedron

no code implementations5 Dec 2013 Nir Ailon, Kohei Hatano, Eiji Takimoto

Unfortunately, CombBand requires at each step an $n$-by-$n$ matrix permanent approximation to within improved accuracy as $T$ grows, resulting in a total running time that is super linear in $T$, making it impractical for large time horizons.

Cannot find the paper you are looking for? You can Submit a new open access paper.