Author name cluster

Hiroyuki Hanada

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

2 author rows

TMLR Journal 2025 Journal Article

Conditional Latent Space Molecular Scaffold Optimization for Accelerated Molecular Design

Onur Boyar
Hiroyuki Hanada
Ichiro Takeuchi

The rapid discovery of new chemical compounds is essential for advancing global health and developing treatments. While generative models show promise in creating novel molecules, challenges remain in ensuring the real-world applicability of these molecules and finding such molecules efficiently. To address this challenge, we introduce Conditional Latent Space Molecular Scaffold Optimization (CLaSMO), which integrates a Conditional Variational Autoencoder (CVAE) with Latent Space Bayesian Optimization (LSBO) to strategically modify molecules while preserving similarity to the original input, effectively framing the task as constrained optimization. Our LSBO setting improves the sample-efficiency of the molecular optimization, and our modification approach helps us to obtain molecules with higher chances of real-world applicability. CLaSMO explores substructures of molecules in a sample-efficient manner by performing BO in the latent space of a CVAE conditioned on the atomic environment of the molecule to be optimized. Our extensive evaluations across diverse optimization tasks—including rediscovery, docking score, and multi‑property optimization—show that CLaSMO efficiently enhances target properties, delivers remarkable sample-efficiency crucial for resource‑limited applications while considering molecular similarity constraints, achieves state of the art performance, and maintains practical synthetic accessibility. We also provide an open-source web application that enables chemical experts to apply CLaSMO in a Human-in-the-Loop setting.

PDF Details

ICML Conference 2025 Conference Paper

Distributionally Robust Active Learning for Gaussian Process Regression

Shion Takeno
Yoshito Okura
Yu Inatsu
Tatsuya Aoyama
Tomonari Tanaka
Satoshi Akahane
Hiroyuki Hanada
Noriaki Hashimoto

Gaussian process regression (GPR) or kernel ridge regression is a widely used and powerful tool for nonlinear prediction. Therefore, active learning (AL) for GPR, which actively collects data labels to achieve an accurate prediction with fewer data labels, is an important problem. However, existing AL methods do not theoretically guarantee prediction accuracy for target distribution. Furthermore, as discussed in the distributionally robust learning literature, specifying the target distribution is often difficult. Thus, this paper proposes two AL methods that effectively reduce the worst-case expected error for GPR, which is the worst-case expectation in target distribution candidates. We show an upper bound of the worst-case expected squared error, which suggests that the error will be arbitrarily small by a finite number of data labels under mild conditions. Finally, we demonstrate the effectiveness of the proposed methods through synthetic and real-world datasets.

Details

TMLR Journal 2025 Journal Article

Distributionally Robust Coreset Selection under Covariate Shift

Tomonari Tanaka
Hiroyuki Hanada
Hanting Yang
Aoyama Tatsuya
Yu Inatsu
Akahane Satoshi
Yoshito Okura
Noriaki Hashimoto

Coreset selection, which involves selecting a small subset from an existing training dataset, is an approach to reducing training data, and various approaches have been proposed for this method. In practical situations where these methods are employed, it is often the case that the data distributions differ between the development phase and the deployment phase, with the latter being unknown. Thus, it is challenging to select an effective subset of training data that performs well across all deployment scenarios. We therefore propose Distributionally Robust Coreset Selection (DRCS). DRCS theoretically derives an estimate of the upper bound for the worst-case test error, assuming that the future covariate distribution may deviate within a defined range from the training distribution. Furthermore, by selecting instances in a way that suppresses the estimate of the upper bound for the worst-case test error, DRCS achieves distributionally robust training instance selection. This study is primarily applicable to convex training computation, but we demonstrate that it can also be applied to deep learning under appropriate approximations. In this paper, we focus on covariate shift, a type of data distribution shift, and demonstrate the effectiveness of DRCS through experiments.

PDF Details

AAAI Conference 2022 Conference Paper

Fast and More Powerful Selective Inference for Sparse High-Order Interaction Model

Diptesh Das
Vo Nguyen Le Duy
Hiroyuki Hanada
Koji Tsuda
Ichiro Takeuchi

Automated high-stake decision-making, such as medical diagnosis, requires models with high interpretability and reliability. We consider the sparse high-order interaction model as an interpretable and reliable model with a good prediction ability. However, finding statistically significant high-order interactions is challenging because of the intrinsically high dimensionality of the combinatorial effects. Another problem in data-driven modeling is the effect of “cherry-picking” (i. e. , selection bias). Our main contribution is extending the recently developed parametric programming approach for selective inference to high-order interaction models. An exhaustive search over the cherry tree (all possible interactions) can be daunting and impractical, even for small-sized problems. We introduced an efficient pruning strategy and demonstrated the computational efficiency and statistical power of the proposed method using both synthetic and real data.

PDF Details

AAAI Conference 2018 Conference Paper

Efficiently Monitoring Small Data Modification Effect for Large-Scale Learning in Changing Environment

Hiroyuki Hanada
Atsushi Shibagaki
Jun Sakuma
Ichiro Takeuchi

We study large-scale machine learning problems in changing environments where a small part of the dataset is modiﬁed, and the effect of the data modiﬁcation must be monitored in order to know how much the modiﬁcation changes the optimal model. When the entire dataset is large, even if the amount of the data modiﬁcation is fairly small, the computational cost for re-training the model would be prohibitively large. In this paper, we propose a novel method, called the optimal solution bounding (OSB), for monitoring such a data modiﬁcation effect on the optimal model by efﬁciently evaluating (without actually re-training) it. The proposed method provides bounds on the unknown optimal model with the cost proportional only to the size of the data modiﬁcation.

PDF Details

TCS Journal 2014 Journal Article

Average-case linear-time similar substring searching by the q -gram distance

Hiroyuki Hanada
Mineichi Kudo
Atsuyoshi Nakamura

In this paper we consider the problem of similar substring searching in the q-gram distance. The q-gram distance d q ( x, y ) is a similarity measure between two strings x and y defined by the number of different q-grams between them. The distance can be used instead of the edit distance due to its lower computation cost, O ( | x | + | y | ) vs. O ( | x | | y | ), and its good approximation for the edit distance. However, if this distance is applied to the problem of finding all similar strings, in a long text t, to a given pattern p, the total computation cost is sometimes not acceptable. Ukkonen already proposed two fast algorithms: one with an array and the other with a tree. When “similar” means k or less in d q, their time complexities are O ( | t | k + | p | ) and O ( | t | log k + | p | ), respectively. In this paper, we propose two algorithms of average-case complexity O ( | t | + | p | ), although their worst-case complexities are still O ( | t | k + | p | ) and O ( | t | log k + | p | ), respectively. The linearity of the average-case complexity is analyzed under the assumption of random sampling of t and the condition that q is larger than a threshold. The algorithms exploit the fact that similar substrings in t are often found at very close positions if the beginning positions of the substrings are close. In the second proposed algorithm, we adopted a doubly-linked list supported by an array and a search tree to search for a list element in O ( log k ) time. Experimental results support their theoretical average-case complexities.

Details DOI