Parallel coordinate plots (PCPs) are a prevalent method to interpret the relationship between the control parameters and metrics. PCPs deliver such an interpretation by color gradation based on a single metric. However, it is challenging to provide such a gradation when multiple metrics are present. Although a naïve approach involves calculating a single metric by linearly weighting each metric, such weighting is unclear for users. To address this problem, we first propose a principled formulation for calculating the optimal weight based on a specific preferred metric combination. Although users can simply select their preference from a two-dimensional (2D) plane for bimetric problems, multi-metric problems require intuitive visualization to allow them to select their preference. We achieved this using various radar charts to visualize the metric trade-offs on the 2D plane reduced by UMAP. In the analysis using pedestrian flow guidance planning, our method identified unique patterns of control parameter importance for each user preference, highlighting the effectiveness of our method.

Meetings are important activities that influence the achievement of team objectives. The characteristics of creative meetings must be clarified to enhance creativity. This study conducted two analyses with the aim of identifying participant behaviors that contribute to creativity by quantitatively analyzing creativity in actual meeting data. First, utterances in meetings were annotated using 11 categories, and frequently occurring dialogue patterns were examined both overall and for each team through pattern analysis. Second, regression analysis was conducted to examine the relationship between the frequency of these frequent patterns and a quantitative creativity index constructed based on text sentiment. In the first analysis, the most frequently observed type of interaction in meetings was the exchange of opinions. Additionally, communication characteristics specific to each team were identified. However, no significant relationships were found between the creativity index and the dialogue patterns in the second analysis. Further analysis, including a reconsideration of the method of aggregating conversation patterns, is required in future work.

To design the “Kansei value” aspect of a product, it is useful to design multilayered relationships of perceptual and affective responses via the physical or psychophysical properties of the product. However, because they are qualitative and ambiguous, designing a model is time-consuming. Moreover, the design was conducted by hypothesis and trial-and-error by the experimenter. In this article, we developed a method to automatically construct several semioptimal structures by applying a genetic algorithm to model design based on structural equation modeling, using the results of image measurement and subjective evaluation experiments on various material samples. Under set convergence conditions, the method constructed statistically optimized structures that represent the relationships among adjectives describing perception and affective, and the properties. A semantic validation was performed to determine the final model. As a result, the proposed method could be used to construct a model that can be interpreted as semantically and statistically superior compared to methods in related studies. A unique feature of this article was the use of the physical and psychophysical properties obtained by measurements in the construction of a multilayer model. Also, the advantage of this method is that it can be used to construct important structures that may be overlooked.

Hyperparameter optimization (HPO) is a vital step in improving performance in deep learning (DL). Practitioners are often faced with the trade-off between multiple criteria, such as accuracy and latency. Given the high computational needs of DL and the growing demand for efficient HPO, the acceleration of multi-objective (MO) optimization becomes ever more important. Despite the significant body of work on meta-learning for HPO, existing methods are inapplicable to MO tree-structured Parzen estimator (MO-TPE), a simple yet powerful MO-HPO algorithm. In this paper, we extend TPE’s acquisition function to the meta-learning setting using a task similarity defined by the overlap of top domains between tasks. We also theoretically analyze and address the limitations of our task similarity. In the experiments, we demonstrate that our method speeds up MO-TPE on tabular HPO benchmarks and attains state-of-the-art performance. Our method was also validated externally by winning the AutoML 2022 competition on “Multiobjective Hyperparameter Optimization for Transformers”. See https://arxiv.org/abs/2212.06751 for the latest version with Appendix.

The recent rise in popularity of Hyperparameter Optimization (HPO) for deep learning has highlighted the role that good hyperparameter (HP) space design can play in training strong models. In turn, designing a good HP space is critically dependent on understanding the role of different HPs. This motivates research on HP Importance (HPI), e.g., with the popular method of functional ANOVA (f-ANOVA). However, the original f-ANOVA formulation is inapplicable to the subspaces most relevant to algorithm designers, such as those defined by top performance. To overcome this issue, we derive a novel formulation of f-ANOVA for arbitrary subspaces and propose an algorithm that uses Pearson divergence (PED) to enable a closed-form calculation of HPI. We demonstrate that this new algorithm, dubbed PED-ANOVA, is able to successfully identify important HPs in different subspaces while also being extremely computationally efficient. See https://arxiv.org/abs/2304.10255 for the latest version with Appendix.

Hyperparameter optimization (HPO) is crucial for strong performance of deep learning algorithms and real-world applications often impose some constraints, such as memory usage, or latency on top of the performance requirement. In this work, we propose constrained TPE (c-TPE), an extension of the widely-used versatile Bayesian optimization method, tree-structured Parzen estimator (TPE), to handle these constraints. Our proposed extension goes beyond a simple combination of an existing acquisition function and the original TPE, and instead includes modifications that address issues that cause poor performance. We thoroughly analyze these modifications both empirically and theoretically, providing insights into how they effectively overcome these challenges. In the experiments, we demonstrate that c-TPE exhibits the best average rank performance among existing methods with statistical significance on 81 expensive HPO with inequality constraints. Due to the lack of baselines, we only discuss the applicability of our method to hard-constrained optimization in Appendix D. See https://arxiv.org/abs/2211.14411 for the latest version with Appendix.

To differentiate from competitors, it has become increasingly important to design products based on the “Kansei value,” which impresses and inspires consumers. However, the perceptual indices of products are generally designed qualitatively, which is not only time-consuming but also costly. Therefore, in recent years, studies have been conducted to discuss the relationship between some perceptions by applying multivariate analysis and machine learning to the adjectives related to perception and affective response. Nonetheless, determining the structure that expresses the relationship of the adjectives is obtained by trial and error, based on the hypothesis of the researchers. Therefore, we aimed to investigate a method for mechanically exploring some semi-optimal structures without a hypothesis of the model structure by applying a genetic algorithm to the construction of adjective relationships using structural equation modeling. In this study, we prepared a four-layered model according to the human perception process for eight categories of material samples and constructed a relationship from a total of 20 words perceived from each sample. Consequently, we could construct a perceptual model that can be interpreted quantitatively and semantically using the proposed method. The advantage of this technique is that it can be used to construct important structures that might be overlooked.

Practitioners often encounter challenging real-world problems that involve a simultaneous optimization of multiple objectives in a complex search space. To address these problems, we propose a practical multiobjective Bayesian optimization algorithm. It is an extension of the widely used Tree-structured Parzen Estimator (TPE) algorithm, called Multiobjective Tree-structured Parzen Estimator (MOTPE). We demonstrate that MOTPE approximates the Pareto fronts of a variety of benchmark problems and a convolutional neural network design problem better than existing methods through the numerical results. We also investigate how the configuration of MOTPE affects the behavior and the performance of the method and the effectiveness of asynchronous parallelization of the method based on the empirical results.

Hyperparameter optimization (HPO), formulated as black-box optimization (BBO), is recognized as essential for automation and high performance of machine learning approaches. The CMA-ES is a promising BBO approach with a high degree of parallelism, and has been applied to HPO tasks, often under parallel implementation, and shown superior performance to other approaches including Bayesian optimization (BO). However, if the budget of hyperparameter evaluations is severely limited, which is often the case for end users who do not deserve parallel computing, the CMA-ES exhausts the budget without improving the performance due to its long adaptation phase, resulting in being outperformed by BO approaches. To address this issue, we propose to transfer prior knowledge on similar HPO tasks through the initialization of the CMA-ES, leading to significantly shortening the adaptation time. The knowledge transfer is designed based on the novel definition of task similarity, with which the correlation of the performance of the proposed approach is confirmed on synthetic problems. The proposed warm starting CMA-ES, called WS-CMA-ES, is applied to different HPO tasks where some prior knowledge is available, showing its superior performance over the original CMA-ES as well as BO approaches with or without using the prior knowledge.

Practitioners often encounter computationally expensive multiobjective optimization problems to be solved in a variety of real-world applications. On the purpose of challenging these problems, we propose a new surrogate-based multiobjective optimization algorithm that does not require a large evaluation budget. It is called Multiobjective Tree-structured Parzen Estimator (MOTPE) and is an extension of the tree-structured Parzen estimator widely used to solve expensive single-objective optimization problems. Our empirical evidences reveal that MOTPE can approximate Pareto fronts of many benchmark problems better than existing methods with a limited budget. In this paper, we discuss furthermore the influence of MOTPE configurations to understand its behavior.

In deep learning, hyperparameters can severely affect the learning model performance. The Nelder-Mead (NM) method is known for showing a superior performance for hyperparameter optimization in deep learning. An initial simplex, one of the initial NM method's values, is usually determined randomly while the search performance strongly depends on the shape of the initial simplex. Therefore, it is necessary to determine a proper initial simplex as previous researchers have proposed methods to construct an initial simplex from one starting point in the bounded search space. In this study, we verified how these methods for constructing an initial simplex contribute to improving the result of hyperparameter optimization in deep learning, by using a simple model and a complicated model. A smaller initial simplex may fail to optimization by bad local minima because there are some bad local minima in both learning models. We concluded that the starting point is not necessarily located close to the origin, and that a larger initial simplex contributes to improving the result of hyperparameter optimization in deep learning.