Topic modeling is widely used to uncover latent semantic topics from a corpus. However, topic models often struggle to identify minor topics due to their tendency to prioritize dominant patterns in the data. They are also hindered by polysemous words and general terms, which frequently appear in multiple contexts, making topic assignment difficult. Seed-guided topic modeling addresses these issues by incorporating prior knowledge through “seed words”. Existing approaches, however, primarily rely on supervised selection using label-dependent metrics or manual selection. Both are limited by scalability and susceptible to human bias, particularly when dealing with unstructured real-world data. As a result, the selection of seed words in unsupervised settings remains underexplored. To address these challenges, we propose an automated seed word selection process that identifies diverse and cohesive word sets based on inter-word relationships. We instantiate this process with$\text{SeedCapture}$, an algorithm that utilizes co-occurrence to capture meaningful word associations. Unlike prior methods,$\text{SeedCapture}$operates in a fully unsupervised manner, requiring no predefined labels or human intervention.$\text{SeedCapture}$requires minimal parameter tuning and is highly adaptable, enabling seamless integration into existing seed-guided topic models. Through extensive quantitative and qualitative evaluations across multiple datasets and topic models, we demonstrate that$\text{SeedCapture}$achieves results comparable to those obtained through supervised seed word selection.

Tracking an object in a noisy environment is difficult, especially when unknown parameters affect the object’s behavior. In the case of a high-speed ballistic object, its trajectory is affected by changes in atmospheric conditions as well as various parameters of the object itself. To filter these latent factors of the dynamics model, this paper proposes a black-box expectation–maximization algorithm to estimate the latent parameters and enhance the accuracy of object tracking. The expectation step involves calculating the likelihood of observations through extended Kalman smoothing, which reflects the forward–backward probability combination. The maximization step involves optimizing the unknown parameter, which is the object mass, to maximize the likelihood through Bayesian optimization with Gaussian process regressions. Our simulation experiments show that our algorithm reduces the error of tracking a ballistic object’s position given noisy observations with unknown parameters.

Topic modeling is widely used to uncover latent semantic topics from a corpus. However, topic models often struggle to identify minor topics due to their tendency to prioritize dominant patterns in the data. They are also hindered by polysemous words and general terms, which frequently appear in multiple contexts, making topic assignment difficult. Seed-guided topic modeling addresses these issues by incorporating prior knowledge through “seed words”. Existing approaches, however, primarily rely on supervised selection using label-dependent metrics or manual selection. Both are limited by scalability and susceptible to human bias, particularly when dealing with unstructured real-world data. As a result, the selection of seed words in unsupervised settings remains underexplored. To address these challenges, we propose an automated seed word selection process that identifies diverse and cohesive word sets based on inter-word relationships. We instantiate this process with$\text{SeedCapture}$, an algorithm that utilizes co-occurrence to capture meaningful word associations. Unlike prior methods,$\text{SeedCapture}$operates in a fully unsupervised manner, requiring no predefined labels or human intervention.$\text{SeedCapture}$requires minimal parameter tuning and is highly adaptable, enabling seamless integration into existing seed-guided topic models. Through extensive quantitative and qualitative evaluations across multiple datasets and topic models, we demonstrate that$\text{SeedCapture}$achieves results comparable to those obtained through supervised seed word selection.

Tracking an object in a noisy environment is difficult, especially when unknown parameters affect the object’s behavior. In the case of a high-speed ballistic object, its trajectory is affected by changes in atmospheric conditions as well as various parameters of the object itself. To filter these latent factors of the dynamics model, this paper proposes a black-box expectation–maximization algorithm to estimate the latent parameters and enhance the accuracy of object tracking. The expectation step involves calculating the likelihood of observations through extended Kalman smoothing, which reflects the forward–backward probability combination. The maximization step involves optimizing the unknown parameter, which is the object mass, to maximize the likelihood through Bayesian optimization with Gaussian process regressions. Our simulation experiments show that our algorithm reduces the error of tracking a ballistic object’s position given noisy observations with unknown parameters.

Active learning effectively collects data instances for training deep learning models when the labeled dataset is limited and the annotation cost is high. Besides active learning, data augmentation is also an effective technique to enlarge the limited amount of labeled instances. However, the potential gain from virtual instances generated by data augmentation has not been considered in the acquisition process of active learning yet. Looking ahead the effect of data augmentation in the process of acquisition would select and generate the data instances that are informative for training the model. Hence, this paper proposes Look-Ahead Data Acquisition via augmentation, or LADA, to integrate data acquisition and data augmentation. LADA considers both 1) unlabeled data instance to be selected and 2) virtual data instance to be generated by data augmentation, in advance of the acquisition process. Moreover, to enhance the informativeness of the virtual data instances, LADA optimizes the data augmentation policy to maximize the predictive acquisition score, resulting in the proposal of InfoMixup and InfoSTN. As LADA is a generalizable framework, we experiment with the various combinations of acquisition and augmentation methods. The performance of LADA shows a significant improvement over the recent augmentation and acquisition baselines which were independently applied to the benchmark datasets.

In order to estimate the launch point of a high-speed vehicle, predicting the various characteristics of the vehicle's movement, such as drag and thrust, must be preceded by the estimation. To predict the various parameters regarding the vehicle's characteristics, we build the IMM filter specialized in predicting the parameters of the post-launch phase based on flight dynamics. Then we estimate the launch point of the high-speed vehicle using Inverse Dynamics. In addition, we assume the arbitrary error level of the radar for accuracy of the prediction. We organize multiple-dimensioned IMM structures, and figure out the optimal value of parameters by comparing the various IMM structures. After deriving the optimal value of parameters, we verify the launch point estimation error under certain error level.

This study proposes a novel Debiased Active Learning (DAL) approach for solar photovoltaic (PV) power prediction based on ground-based sky images. Conventional uncertainty-based sampling methods often suffer from selection bias, particularly under class-imbalanced conditions where clear-sky samples dominate over cloudy-sky samples. To address this issue, the proposed DAL method estimates a debiasing matrix from a small, trusted validation dataset and uses it to correct the model’s predictive probabilities before uncertainty sampling. Using the SKIPP’D (Sky Images and Photovoltaic Power Dataset) dataset(≈300,000 paired sky images and PV power values), the proposed DAL significantly improves labeling efficiency and prediction accuracy compared to traditional uncertainty-based active learning approaches.

Recent studies identified that sequential Recommendation is improved by the attention mechanism. By following this development, we propose Relation-Aware Kernelized Self-Attention (RKSA) adopting a self-attention mechanism of the Transformer with augmentation of a probabilistic model. The original self-attention of Transformer is a deterministic measure without relation-awareness. Therefore, we introduce a latent space to the self-attention, and the latent space models the recommendation context from relation as a multivariate skew-normal distribution with a kernelized covariance matrix from co-occurrences, item characteristics, and user information. This work merges the self-attention of the Transformer and the sequential recommendation by adding a probabilistic model of the recommendation task specifics. We experimented RKSA over the benchmark datasets, and RKSA shows significant improvements compared to the recent baseline models. Also, RKSA were able to produce a latent space model that answers the reasons for recommendation.