arXiv subjects

Hisashi Kashima

Publications and source records attributed to Hisashi Kashima.

80 records · Page 5Linked to original sources

arXiv2018

Dual Convolutional Neural Network for Graph of Graphs Link Prediction

Graphs are general and powerful data representations which can model complex real-world phenomena, ranging from chemical compounds to social networks; however, effective feature extraction from graphs is not a trivial task, and much work has been done in the field of machine learning and data mining. The recent advances in graph neural networks have made automatic and flexible feature extraction from graphs possible and have improved the predictive performance significantly. In this paper, we go further with this line of research and address a more general problem of learning with a graph of graphs (GoG) consisting of an external graph and internal graphs, where each node in the external graph has an internal graph structure. We propose a dual convolutional neural network that extracts node representations by combining the external and internal graph structures in an end-to-end manner. Experiments on link prediction tasks using several chemical network datasets demonstrate the effectiveness of the proposed method.

cs.LG↗

arXiv2018

BayesGrad: Explaining Predictions of Graph Convolutional Networks

Recent advances in graph convolutional networks have significantly improved the performance of chemical predictions, raising a new research question: "how do we explain the predictions of graph convolutional networks?" A possible approach to answer this question is to visualize evidence substructures responsible for the predictions. For chemical property prediction tasks, the sample size of the training data is often small and/or a label imbalance problem occurs, where a few samples belong to a single class and the majority of samples belong to the other classes. This can lead to uncertainty related to the learned parameters of the machine learning model. To address this uncertainty, we propose BayesGrad, utilizing the Bayesian predictive distribution, to define the importance of each node in an input graph, which is computed efficiently using the dropout technique. We demonstrate that BayesGrad successfully visualizes the substructures responsible for the label prediction in the artificial experiment, even when the sample size is small. Furthermore, we use a real dataset to evaluate the effectiveness of the visualization. The basic idea of BayesGrad is not limited to graph-structured data and can be applied to other data types.

cs.LG↗

arXiv2017

Matrix- and tensor-based recommender systems for the discovery of currently unknown inorganic compounds

Chemically relevant compositions (CRCs) and atomic arrangements of inorganic compounds have been collected as inorganic crystal structure databases. Machine-learning is a unique approach to search for currently unknown CRCs from vast candidates. Herein we propose matrix- and tensor-based recommender system approaches to predict currently unknown CRCs from database entries of CRCs. Firstly, the performance of the recommender system approaches to discover currently unknown CRCs is examined. A Tucker decomposition recommender system shows the best discovery rate of CRCs as the majority of the top 100 recommended ternary and quaternary compositions correspond to CRCs. Secondly, systematic density functional theory (DFT) calculations are performed to investigate the phase stability of the recommended currently unknown CRCs. The phase stability of the 27 compositions reveals that 23 currently unknown compounds are stable. These results indicate that the recommender system has great potential to accelerate the discovery of new compounds.

cond-mat.mtrl-sci↗

arXiv2017

Using Posters to Recommend Anime and Mangas in a Cold-Start Scenario

Item cold-start is a classical issue in recommender systems that affects anime and manga recommendations as well. This problem can be framed as follows: how to predict whether a user will like a manga that received few ratings from the community? Content-based techniques can alleviate this issue but require extra information, that is usually expensive to gather. In this paper, we use a deep learning technique, Illustration2Vec, to easily extract tag information from the manga and anime posters (e.g., sword, or ponytail). We propose BALSE (Blended Alternate Least Squares with Explanation), a new model for collaborative filtering, that benefits from this extra information to recommend mangas. We show, using real data from an online manga recommender system called Mangaki, that our model improves substantially the quality of recommendations, especially for less-known manga, and is able to provide an interpretation of the taste of the users.

cs.IR↗

arXiv2015

Regret Lower Bound and Optimal Algorithm in Dueling Bandit Problem

We study the $K$-armed dueling bandit problem, a variation of the standard stochastic bandit problem where the feedback is limited to relative comparisons of a pair of arms. We introduce a tight asymptotic regret lower bound that is based on the information divergence. An algorithm that is inspired by the Deterministic Minimum Empirical Divergence algorithm (Honda and Takemura, 2010) is proposed, and its regret is analyzed. The proposed algorithm is found to be the first one with a regret upper bound that matches the lower bound. Experimental comparisons of dueling bandit algorithms show that the proposed algorithm significantly outperforms existing ones.

stat.ML↗

arXiv2012

Fast Computation of Subpath Kernel for Trees

The kernel method is a potential approach to analyzing structured data such as sequences, trees, and graphs; however, unordered trees have not been investigated extensively. Kimura et al. (2011) proposed a kernel function for unordered trees on the basis of their subpaths, which are vertical substructures of trees responsible for hierarchical information in them. Their kernel exhibits practically good performance in terms of accuracy and speed; however, linear-time computation is not guaranteed theoretically, unlike the case of the other unordered tree kernel proposed by Vishwanathan and Smola (2003). In this paper, we propose a theoretically guaranteed linear-time kernel computation algorithm that is practically fast, and we present an efficient prediction algorithm whose running time depends only on the size of the input tree. Experimental results show that the proposed algorithms are quite efficient in practice.

cs.DS↗

arXiv2012

Parametric Return Density Estimation for Reinforcement Learning

Most conventional Reinforcement Learning (RL) algorithms aim to optimize decision-making rules in terms of the expected returns. However, especially for risk management purposes, other risk-sensitive criteria such as the value-at-risk or the expected shortfall are sometimes preferred in real applications. Here, we describe a parametric method for estimating density of the returns, which allows us to handle various criteria in a unified manner. We first extend the Bellman equation for the conditional expected return to cover a conditional probability density of the returns. Then we derive an extension of the TD-learning algorithm for estimating the return densities in an unknown environment. As test instances, several parametric density estimation algorithms are presented for the Gaussian, Laplace, and skewed Laplace distributions. We show that these algorithms lead to risk-sensitive as well as robust RL paradigms through numerical experiments.

cs.LG↗

arXiv2011

Estimation of low-rank tensors via convex optimization

In this paper, we propose three approaches for the estimation of the Tucker decomposition of multi-way arrays (tensors) from partial observations. All approaches are formulated as convex minimization problems. Therefore, the minimum is guaranteed to be unique. The proposed approaches can automatically estimate the number of factors (rank) through the optimization. Thus, there is no need to specify the rank beforehand. The key technique we employ is the trace norm regularization, which is a popular approach for the estimation of low-rank matrices. In addition, we propose a simple heuristic to improve the interpretability of the obtained factorization. The advantages and disadvantages of three proposed approaches are demonstrated through numerical experiments on both synthetic and real world datasets. We show that the proposed convex optimization based approaches are more accurate in predictive performance, faster, and more reliable in recovering a known multilinear structure than conventional approaches.

stat.ML↗