Round rl-randomized-lavaworld-aug2023-train Train DatasetThis is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of Reinforcement Learning agents trained to navigate the Lavaworld Minigrid environment. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers.

nlp-question-answering-aug2023-trainThis is the train data used to evaluate trojan detection software solutions. This data, generated at NIST, consists of natural language processing (NLP) AIs trained to perform extractive question answering on English text. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers.

A table of known interferences for per- and polyfluoroalkyl substances using mass spectrometry.

Round rl-lavaworld-jul2023-train Train DatasetThis is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of Reinforcement Learning agents trained to navigate the Lavaworld Minigrid environment. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers.

Data here contain and describe an open-source structured query language (SQLite) portable database containing high resolution mass spectrometry data (MS1 and MS2) for per- and polyfluorinated alykl substances (PFAS) and associated metadata regarding their measurement techniques, quality assurance metrics, and the samples from which they were produced. These data are stored in a format adhering to the Database Infrastructure for Mass Spectrometry (DIMSpec) project. That project produces and uses databases like this one, providing a complete toolkit for non-targeted analysis.

This data publication contains the mass spectrometry chemical characterization of microplastic and nanoplastic chemical analysis. The data from this study includes mass spectra of pure, mixed, and weathered microplastics and nanoplastics at high and low fragmentation, extracted ion chronograms, Kendrick mass defect plots, code, and the derived and processed data. The data analysis code (MATLAB 2022a*) used for unsupervised learning of cluster and compositional relationships is also included.

Round 13 Train DatasetThis is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of object detection AIs trained both on synthetic image data build from Cityscapes and the DOTA_v2 dataset. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers. This dataset consists of 128 AI models using a small set of model architectures.

Round 12 Train DatasetThis is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of pdf malware classification AIs trained Contaigio dataset feature vectors. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers. This dataset consists of 120 AI models using a small set of model architectures.

Round 11 Train DatasetThis is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of image classification AIs trained on synthetic image data build from Cityscapes. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers. This dataset consists of 288 AI models using a small set of model architectures.

Python scripts and Python+Qt graphical user interface for calculating Feldman-Cousins confidence intervals for low-count Poisson processes in the presence of a known background and for Gaussian processes with a physical lower limit of 0.