We design Big Data infrastructures to enable efficient, open and reproducible neuroinformatics.
New pre-print: Dynamic Ensemble Size Adjustment for Memory Constrained Mondrian Forest
New pre-print: Sea: A lightweight data-placement library for Big Data scientific computing
New pre-print: NeuroCI: Continuous Integration of Neuroimaging Results Across Software Pipelines and Datasets
New pre-print: Mondrian Forest for Data Stream Classification Under Memory Constraints
🚨 Open position: postdoctoral fellow (now filled)
New pre-print: PyTracer: Automatically profiling numerical instabilities in Python
New pre-print: The benefits of prefetching for large-scale cloud-based neuroimaging analysis workflows
New pre-print: A Recommender System for Scientific Datasets and Analysis Pipelines
New pre-print: Accurate simulation of operating system updates in neuroimaging using Monte-Carlo arithmetic
New pre-print: Reducing numerical precision preserves classification accuracy in Mondrian Forests
New pre-print: Modeling the Linux page cache for accurate simulation of data-intensive applications
New pre-print: Data Augmentation Through Monte Carlo Arithmetic Leads to More Generalizable Classification in Connectomics
New pre-print: An Analysis of Security Vulnerabilities in Container Images for Scientific Data Analysis
New pre-print: Numerical Instabilities in Analytical Pipelines Lead to Large and Meaningful Variability in Brain Networks.
New pre-print: A benchmark of data stream classification for human activity recognition on connected objects.
New pre-print: Can we Estimate Truck Accident Risk from Telemetric Data using Machine Learning?
New pre-print: File-based localization of numerical perturbations in data analysis pipelines.
New pre-print: Performance benefits of Intel® Optane™ DC persistent memory for the parallel processing of large neuroimaging data.