Recent rapid advancements of machine learning have greatly enhanced the accuracy of prediction models, but most models remain "black boxes", making prediction error diagnosis challenging, especially with outliers. This lack of transparency hinders trust and reliability in industrial applications. Heuristic attribution methods, while helpful, often fail to capture true causal relationships, leading to inaccurate error attributions. Various root-cause analysis methods have been developed using Shapley values, yet they typically require predefined causal graphs, limiting their applicability for prediction errors in machine learning models. To address these limitations, we introduce the Causal-Discovery-based Root-Cause Analysis (CD-RCA) method that estimates causal relationships between the prediction error and the explanatory variables, without needing a pre-defined causal graph. By simulating synthetic error data, CD-RCA can identify variable contributions to outliers in prediction errors by Shapley values. Extensive experiments show CD-RCA outperforms current heuristic attribution methods.

Low visibility condition affects the safety and operation of road, sea, and air traffic, which causes disruption in transportation and sometimes leads to serious accidents. Therefore, forecasting of low visibility accurately is of utmost importance for transportation agencies and aviation authorities. The conventional approach of visibility forecasting using time-series weather data cannot forecast low visibility with high accuracy due to imbalanced and clipped visibility data used to build a forecasting model. To address this issue, we propose a new approach that combines forecasting and classification methods with discrete and kernel density estimation (KDE)-based weights to build forecasting models for multi-step ahead forecasting. We show the effectiveness of our proposed approach by performing experiments with four publicly available visibility datasets of weather and air quality data from the USA and Japan and evaluating in terms of forecasting errors and accuracy of forecasting of low visibility. Experimental results suggest that our proposed approach can improve RMSE and MAE in the low visibility range by respectively 29–54 % and 40–59 % and F-score of forecasting of low and high visibility by 0.20-0.26 on the average over existing technologies for 3–24 hour ahead forecasting, which implies that it is well-suited for forecasting of low visibility.

Large amounts of time-series data have become accessible due to the rapid development of Internet-of-Things technologies and the demand for extracting useful knowledge from these data is increasing. Toward this goal, time-series segmentation — dividing data into similar segments — is a promising method for understanding the mechanisms in time-series data. In this paper, we focus on time-lag that appears in real datasets. Time lag — a typical phenomenon in time-series data — occurs when the speed of information diffusion differs between variables. However, conventional methods cannot distinguish differences in segmentation positions. In response, we propose Lag-Aware Multivariate Time-Series Segmentation (LAMTSS), an algorithm capturing time-lag across variables to determine segmentation positions for each variable. LAMTSS utilizes dynamic time warping without hyperparameter tuning. We confirm the accuracy of LAMTSS using artificial datasets and demonstrate the discovery of useful knowledge in real datasets.

Inspired by the Bayesian brain hypothesis and deep learning, we develop a Bayesian autoencoder, a method of constructing recognition systems using a Bayesian network. We construct hierarchical Bayesian networks based on feature extraction and implement pooling to achieve invariance within a Bayesian network framework. The constructed networks propagate information bidirectionally between layers. We expect they will be able to achieve brain-like recognition using local features and global information such as their environments.

We propose Bayesian AutoEncoder (BAE) in order to construct a recognition system which uses feedback information. BAE constructs a generative model of input data as a Bayes Net. The network trained by BAE obtains its hidden variables as the features of given data. It can execute inference for each variable through belief propagation, using both feedforward and feedback information. We confirmed that BAE can construct small networks with one hidden layer and extract features as hidden variables from 3x3 and 5x5 pixel input data.