Abstract
This study investigates the integration of deep learning for single-modality and multimodal data within materials science. Traditional methods for materials discovery are often resource-intensive and slow, prompting the exploration of machine learning to streamline the prediction of material properties. While single-modality models have been effective, they often miss the complexities inherent in material data. The paper explores multimodal data integration—combining text, images, and tabular data—and demonstrates its potential to improve predictive accuracy. Utilizing the Alexandria dataset, the research introduces a custom methodology involving multimodal data creation, model tuning with AutoGluon framework, and evaluation through targeted fusion techniques. Results reveal that multimodal approaches enhance predictive accuracy and efficiency, particularly when text and image data are integrated. However, challenges remain in predicting complex features like band gaps. Future directions include incorporating new data types and refining specialized models to improve materials discovery and innovation.

Abstract
Accurate retail demand forecasting is essential for optimizing operations, improving customer satisfaction, and enhancing financial performance. Traditional statistical models often struggle to handle the complexities of retail time series data, which include hierarchical structures, irregular patterns, and external influencing factors. In this study, we evaluate the effectiveness of various Transformer-based models for probabilistic time series forecasting in retail, leveraging the rich explanatory variables provided by the M5 dataset. The models incorporate diverse features, including calendar-related information, selling prices, and socio-economic indicators such as SNAP activities, to capture the temporal, promotional, and socio-economic dynamics influencing sales. Our results demonstrate that Transformer-based models augmented with explanatory variables outperform their counterparts, providing more accurate and reliable forecasts across different horizons. We show that these models can effectively leverage context to improve forecast accuracy and capture uncertainty through probabilistic forecasting methods. This study highlights the potential of deep learning models in retail demand forecasting and underscores the importance of integrating domain-specific variables to achieve robust, context-aware predictions in dynamic retail environments.

Abstract
This study investigates the effectiveness of different hyperparameter tuning strategies for peer-to-peer risk management. Ensemble learning techniques have shown superior performance in this field compared to individual classifiers and traditional statistical methods. However, model performance is influenced not only by the choice of algorithm but also by hyperparameter tuning, which impacts both predictive accuracy and computational efficiency. This research compares the performance and efficiency of three widely used hyperparameter tuning methods, Grid Search, Random Search, and Optuna, across XGBoost, LightGBM, and Logistic Regression models. The analysis uses the Lending Club dataset, spanning from 2007 Q1 to 2020 Q3, with comprehensive data preprocessing to address missing values, class imbalance, and feature engineering. Model explainability is assessed through feature importance analysis to identify key drivers of default probability. The findings reveal comparable predictive performance among the tuning methods, evaluated using metrics such as G-mean, sensitivity, and specificity. However, Optuna significantly outperforms the others in computational efficiency; for instance, it is 10.7 times faster than Grid Search for XGBoost and 40.5 times faster for LightGBM. Additionally, variations in feature importance rankings across tuning methods influence model interpretability and the prioritization of risk factors. These insights underscore the importance of selecting appropriate hyperparameter tuning strategies to optimize both performance and explainability in peer-to-peer risk management models.