To mitigate the pain of manually tuning hyperparameters of deep neural networks, automated machine learning (AutoML) methods have been developed to search for an optimal set of hyperparameters in large combinatorial search spaces. However, the search results of AutoML methods significantly depend on initial configurations, making it a non-trivial task to find a proper configuration. Therefore, human intervention via a visual analytic approach bears huge potential in this task. In response, we propose HyperTendril, a web-based visual analytics system that supports user-driven hyperparameter tuning processes in a model-agnostic environment. HyperTendril takes a novel approach to effectively steering hyperparameter optimization through an iterative, interactive tuning procedure that allows users to refine the search spaces and the configuration of the AutoML method based on their own insights from given results. Using HyperTendril, users can obtain insights into the complex behaviors of various hyperparameter search algorithms and diagnose their configurations. In addition, HyperTendril supports variable importance analysis to help the users refine their search spaces based on the analysis of relative importance of different hyperparameters and their interaction effects. We present the evaluation demonstrating how HyperTendril helps users steer their tuning processes via a longitudinal user study based on the analysis of interaction logs and in-depth interviews while we deploy our system in a professional industrial environment.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1109/TVCG.2020.3030380 | DOI Listing |
Transition from pediatric to adult care for adolescents and young adults (AYAs) with chronic illness affects the entire family. However, little research has compared AYA and parent experiences of transition. Using Sandelowski and Barroso's method, the aim of this metasynthesis was to summarize findings of qualitative studies focusing on the transition experiences of AYAs and their parents across different chronic physical illnesses.
View Article and Find Full Text PDFJ Chem Inf Model
December 2024
Orion Pharma, Orionintie 1A, 02101 Espoo, Finland.
Given the size of the relevant chemical space for drug discovery, working with fully enumerated compound libraries (especially in three-dimensional (3D)) is unfeasible. Nonenumerated virtual chemical spaces are a practical solution to this issue, where compounds are described as building blocks which are then connected by rules. One concrete example of such is the BioSolveIT chemical spaces file format (.
View Article and Find Full Text PDFNeuropsychol Rev
December 2024
Laboratory of Neuropsychology of Memory, Department of Clinical Neuroscience and Neurorehabilitation, IRCCS Santa Lucia Foundation, Via Ardeatina 306, 00179, Rome, Italy.
To date, few studies have focused on the benefits of dopaminergic treatment on episodic memory functions in patients affected by Parkinson's disease (PD). We conducted a meta-analysis to determine the effects of pharmacological therapy with dopamine in alleviating episodic memory deficits in Parkinson's patients. A secondary aim was to evaluate the role of dopamine in episodic memory circuits and thus in different memory systems.
View Article and Find Full Text PDFWater Res
December 2024
CERIS, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, Lisboa 1049-001, Portugal.
Green walls for greywater treatment have emerged as a solution to increase green spaces in densely urbanized areas while providing treated greywater for reuse. Over the past decade, numerous studies have focused on optimizing these systems, though most address specific operational conditions and evaluate a limited set of performance parameters. This review synthesizes the existing literature using a meta-analysis to identify key operational factors and treatment performance metrics.
View Article and Find Full Text PDFNAR Genom Bioinform
December 2024
Systems Rheumatology, German Rheumatism Research Centre Berlin, Virchowweg 12, 10117 Berlin, Germany.
There is a growing interest in generating bimodal, single-cell RNA sequencing (RNA-seq) data for studying biological pathways. These data are predominantly utilized in understanding phenotypic trajectories using RNA velocities; however, the shape information encoded in the two-dimensional resolution of such data is not yet exploited. In this paper, we present an elliptical parametrization of two-dimensional RNA-seq data, from which we derived statistics that reveal four different modalities.
View Article and Find Full Text PDFEnter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!