arXiv subjects

Masahiro Kato

Publications and source records attributed to Masahiro Kato.

77 records · Page 5Linked to original sources

arXiv2020

Confidence Interval for Off-Policy Evaluation from Dependent Samples via Bandit Algorithm: Approach from Standardized Martingales

This study addresses the problem of off-policy evaluation (OPE) from dependent samples obtained via the bandit algorithm. The goal of OPE is to evaluate a new policy using historical data obtained from behavior policies generated by the bandit algorithm. Because the bandit algorithm updates the policy based on past observations, the samples are not independent and identically distributed (i.i.d.). However, several existing methods for OPE do not take this issue into account and are based on the assumption that samples are i.i.d. In this study, we address this problem by constructing an estimator from a standardized martingale difference sequence. To standardize the sequence, we consider using evaluation data or sample splitting with a two-step estimation. This technique produces an estimator with asymptotic normality without restricting a class of behavior policies. In an experiment, the proposed estimator performs better than existing methods, which assume that the behavior policy converges to a time-invariant policy.

stat.ML↗

arXiv2020

Identifying Different Definitions of Future in the Assessment of Future Economic Conditions: Application of PU Learning and Text Mining

The Economy Watcher Survey, which is a market survey published by the Japanese government, contains \emph{assessments of current and future economic conditions} by people from various fields. Although this survey provides insights regarding economic policy for policymakers, a clear definition of the word "future" in future economic conditions is not provided. Hence, the assessments respondents provide in the survey are simply based on their interpretations of the meaning of "future." This motivated us to reveal the different interpretations of the future in their judgments of future economic conditions by applying weakly supervised learning and text mining. In our research, we separate the assessments of future economic conditions into economic conditions of the near and distant future using learning from positive and unlabeled data (PU learning). Because the dataset includes data from several periods, we devised new architecture to enable neural networks to conduct PU learning based on the idea of multi-task learning to efficiently learn a classifier. Our empirical analysis confirmed that the proposed method could separate the future economic conditions, and we interpreted the classification results to obtain intuitions for policymaking.

econ.EM↗

arXiv2020

Model Specification Test with Unlabeled Data: Approach from Covariate Shift

We propose a novel framework of the model specification test in regression using unlabeled test data. In many cases, we have conducted statistical inferences based on the assumption that we can correctly specify a model. However, it is difficult to confirm whether a model is correctly specified. To overcome this problem, existing works have devised statistical tests for model specification. Existing works have defined a correctly specified model in regression as a model with zero conditional mean of the error term over train data only. Extending the definition in conventional statistical tests, we define a correctly specified model as a model with zero conditional mean of the error term over any distribution of the explanatory variable. This definition is a natural consequence of the orthogonality of the explanatory variable and the error term. If a model does not satisfy this condition, the model might lack robustness with regards to the distribution shift. The proposed method would enable us to reject a misspecified model under our definition. By applying the proposed method, we can obtain a model that predicts the label for the unlabeled test data well without losing the interpretability of the model. In experiments, we show how the proposed method works for synthetic and real-world datasets.

stat.ME↗

arXiv2019

Particle Filter on Episode

Differently from animals, robots can record its experience correctly for long time. We propose a novel algorithm that runs a particle filter on the time sequence of the experience. It can be applied to some teach-and-replay tasks. In a task, the trainer controls a robot, and the robot records its sensor readings and its actions. We name the sequence of the record an episode, which is derived from the episodic memory of animals. After that, the robot executes the particle filter so as to find a similar situation with the current one from the episode. If the robot chooses the action taken in the similar situation, it can replay the taught behavior. We name this algorithm the particle filter on episode (PFoE). The robot with PFoE shows not only a simple replay of a behavior but also recovery motion from skids and interruption. In this paper, we evaluate the properties of PFoE with a small mobile robot.

cs.RO↗

arXiv2018

Alternate Estimation of a Classifier and the Class-Prior from Positive and Unlabeled Data

We consider a problem of learning a binary classifier only from positive data and unlabeled data (PU learning) and estimating the class-prior in unlabeled data under the case-control scenario. Most of the recent methods of PU learning require an estimate of the class-prior probability in unlabeled data, and it is estimated in advance with another method. However, such a two-step approach which first estimates the class prior and then trains a classifier may not be the optimal approach since the estimation error of the class-prior is not taken into account when a classifier is trained. In this paper, we propose a novel unified approach to estimating the class-prior and training a classifier alternately. Our proposed method is simple to implement and computationally efficient. Through experiments, we demonstrate the practical usefulness of the proposed method.

cs.LG↗