Sports scheduling is one of the widely investigated application problems in artificial intelligence and operations research. This problem can be represented as a combinatorial optimization problem, in which the date and the venue of each game must be determined, subject to a given set of constraints. In 2018, Japan Basketball Association (JBA) has started to implement the league games in the prefectures, and regarding U12 (under 12 years old), league games are currently being held in most prefectures. In this paper, the main focus is laid on the U12 Basketball League Scheduling Problem. First, a formal framework for the U12 Basketball League Scheduling based on minimizing the total traveling distance (U12-BLSdist) is defined. A novel solution criterion called an egalitarian solution for U12-BLSdist is also provided. Next, a formal framework for the U12 Basketball league scheduling problem based on minimizing the number of breaks (U12-BLSbreak) is defined. Furthermore, the frameworks U12-BLSdist and U12-BLSbreak are formalized as 0-1 integer programming problems. In the experiments, we use the real data of U12 basketball league games played in Hyogo prefecture in 2018 and solve the U12-BLSdist and U12-BLSbreak problems. An optimal league scheduling and an egalitarian league scheduling for the U12-BLSdist, and the minimal number of breaks for the U12-BLSbreak are reported by comparing with the league scheduling actually used in 2018.

Domain generalization (DG) is a difficult transfer learning problem aiming to learn a generalizable model for unseen domains. Recent foundation models (FMs) are robust to many distribution shifts and, therefore, should substantially improve the performance of DG. In this work, we study generic ways to adopt contrastive languageimage pre-training (CLIP), a visual-language foundation model, for DG problems in image classification. While empirical risk minimization (ERM) greatly improves the accuracy with bigger backbones and training datasets using standard DG benchmarks, fine-tuning FMs is not practical in many real-world situations. We propose Domain Prompt Learning (DPL) as a novel approach for domain inference in the form of conditional prompt generation. DPL achieved a significant accuracy improvement with only training a lightweight prompt generator (a three-layer MLP), whose parameter is of equivalent scale to the classification projector in the previous DG literature. Combining DPL with CLIP provides surprising performance, raising the accuracy of zero-shot CLIP from 73.7% to 79.3% on several standard datasets, namely PACS, VLCS, OfficeHome, and TerraIncognita. We hope the simplicity and success of our approach lead to broader adoption and analysis of foundation models in the domain generalization field.

Many high-performance machine learning models in the real world exhibit the black box problem. This issue is widely recognized as needing output reliability and model transparency. XAI (Explainable AI) represents a research field that addresses this issue. Within XAI, feature attribution methods, which clarify the importance of features irrespective of the task or model type, have become a central focus. Evaluating their efficacy based on empirical evidence is essential when proposing new methods. However, extensive debate exists regarding the properties that importance should be possessed, and a consensus on specific evaluation methods remains elusive. Given this context, many existing studies adopt their evaluation techniques, leading to fragmented discussions. This study aims to ”evaluate the evaluation methods,” focusing mainly on the faithfulness metric, deemed especially significant in evaluation criteria. We conducted empirical experiments related to existing evaluation techniques. The experiments approached the topic from two angles: correlation-based comparative evaluations and property verification using random sequences. In the former experiment, we investigated the correlation between faithfulness evaluation tests using numerous models and feature attribution methods. As a result, we found that very few test combinations exhibited high correlation, and many combinations showed low or no correlation. In the latter experiment, we observed that the measured faithfulness varied depending on the model and dataset by using random sequences instead of feature attribution methods to verify the properties of the faithfulness tests.