Verifiable AI

The Verifiable AI objectives are organized in four open research questions that constitute four dimensions of the grand challenge resulting from the emergent use of AI in safety-critical applications. These four dimensions also represent the natural way to organize the background material on Verifiable AI:

Dependability with AI: how to design and verify dependable and secure systems that include unverifiable AI components?
Dependability of AI: how to verify the dependability and security of AI components themselves (i.e., domain-independent inference engines, as well as knowledge bases either machine learned from data or manually encoded by human domain experts)?
AI for dependability: which AI techniques can be themselves leveraged to automate the design and verification of systems that include unverified AI components?
Meta-AI dependability: which AI techniques can be themselves leveraged to automate the design and verification of AI components (meta-AI)?

A Simple Guide to Verifiable AI

Tutorial

Low-code platform for the development of safe and trustable software modules for autonomous driving

Tutorial on the DIGIHALL Platform based on Papyrus for Robotics, for the design of vehicle software architectures and for the validation of behavior-tree based scenarios and autonomous driving policies in CARLA/ROS 2

Docker container

Hippodrone

A complete autonomous drone simulation

Docker container

COOLAPS Application Base

This repository contains the base components necessary to bootstrap your own Constraint Object-Oriented Logic Action Programming as a Service (COOLAPS) application docker container.

Jupyter Notebook

AI REGIO Predictive Maintenance using Machine Learning

Data preprocessing, Feature Engineering, Model Development, Evaluation & Selection for Predictive Maintenance in Manufacturing. Part of the development of this asset was supported by the AI REGIO project, which has received funding from the European U...

Docker container

VERIFAI

Verifiable and Explainable RIsk Forecasting Artificial Intelligence Framework

Docker container

Fraunhofer uncertainty metrics for classification tasks

Dockerized AI4EU Acumos component for uncertainty estimation for classification networks

ML Model

Platform-as-a-service for accountable evidential transactions

In the domains of aeronautics, automotive, energy, manufacturing and retail, Munich Innovation Hub for Applied AI proposes novel solutions to counter the complexity and dependability challenges resulting from distributed accountability, the need for more ...

GEOSKOP SLU

As a Service

New generation of 6-month global forecast

Introducing ESFA (Empirical Season Forecast for Agriculture) – your reliable companion in long-term agricultural planning. Our validated AI algorithms, which outperform other state-of-the-art climate prediction models, are at the core of ESFA's capabiliti...

Dataset

Adaptive Test Case Selection - Datasets

Data sets to evaluate test case prioritization and selection, finding test cases likely to fail.

Jupyter Notebook

ProGAP: Progressive Graph Neural Networks with Differential Privacy Guarantees

A novel progressive training algorithm for learning Graph Neural Networks with Differential Privacy guarantees

As a Service

AI REGIO-S5 Enterprise Big Data Analytics Suite: Manufacturing

Descriptive and predictive analytics for manufacturing legacy and operational data related to maintenance, delivering insights on critical manufacturing operations.

Jupyter Notebook

GAP: Differentially Private Graph Neural Networks with Aggregation Perturbation

A tailored Graph Neural Network architecture with Differential Privacy guarantees for both training and inference.

View more AI Assets