8 docs tagged with "Release 2024-12"

The Camera Traps application is both a simulator and IoT device software for utilizing machine learning on the edge in field research. The first implementation specializes in applying computer vision (detection and classification) to wildlife images for animal ecology studies. Two operational modes are supported: "simulation" mode and "demo" mode. When executed in simulation mode, the software serves as a test bed for studying ML models, protocols and techniques that optimize storage, execution time, power and accuracy. It requires an input dataset of images to act as the images that would be generated an IoT camera device; it uses these images to drive the simulation.

This repo provides instructions for extracting workload information from AI-Driven Animal Ecology (ADAE) studies.

The Cyberinfrastructure Knowledge Network (CKN) is an extensible and portable distributed framework designed to optimize AI at the edge—particularly in dynamic environments where workloads may change suddenly (for example, in response to motion detection). CKN enhances edge–cloud collaboration by using historical data, graph representations, and adaptable deployment of AI models to satisfy changing accuracy‑and‑latency demands on edge devices.

FastKG is a knowledge graph embedding training library. Knowledge Graph (KG) embeddings are a way to represent entities and relationships from a KG in a continuous vector space, enabling tasks like link prediction and reasoning. TransE, a popular model, represents relationships as translations in the embedding space, such that for a valid triplet (head, relation, tail), the embedding of the head plus the relation vector is close to the embedding of the tail. Training data for TransE is typically stored in a tab-separated values (TSV) format, where each line represents a triplet, e.g., entity1\trelation1\tentity2. For example a dummy train.tsv should look like this:

The Patra Knowledge Base is a system designed to manage and track AI/ML models, with the objective of making them more accountable and trustworthy. It's a key part of the Patra ModelCards framework, which aims to improve transparency and accountability in AI/ML models throughout their entire lifecycle. This includes the model's initial training phase, subsequent deployments, and ongoing usage, whether by the same or different individuals.

The Patra Toolkit is a component of the Patra ModelCards framework designed to simplify the process of creating and documenting AI/ML models. It provides a structured schema that guides users in providing essential information about their models, including details about the model's purpose, development process, and performance. The toolkit also includes features for semi-automating the capture of key information, such as fairness and explainability metrics, through integrated analysis tools. By reducing the manual effort involved in creating model cards, the Patra Toolkit encourages researchers and developers to adopt best practices for documenting their models, ultimately contributing to greater transparency and accountability in AI/ML development.

Language agents for data-driven scientific discovery tasks. The agent features the self-debug mechanism and shows the best performance on ScienceAgentBench.

GitHub Repo