Sign in
Ctrl K

DIANNA - Deep Insight and Neural Networks Analysis

Explainable AI tool for scientists

Modern scientific challenges are often tackled with (Deep) Neural Networks (DNN). Despite their high predictive accuracy, DNNs lack inherent explainability. Many scientists do not harvest DNNs power because of lack of trust and understanding of their working. Meanwhile, the eXplainable AI (XAI) research offers some post-hoc (after training) interpretability methods that provide insight into the DNN reasoning by quantifying the relevance of individual features (image pixels, words in text, etc.) concerning the prediction. These relevance heatmaps indicate how the network has reached its decision directly in the input modality (images, text, speech etc.) of the scientific data. Representing visually the captured knowledge by the AI system can become a source of scientific insights. There are many Open Source Software (OSS) implementations of these methods, alas, supporting a single DNN format, while standards like Open Neural Network eXchange (ONNX) exist. The libraries are known mostly by the AI experts. For the adoption by the wide scientific community understanding of the XAI methods and well-documented and standardized OSS are needed. The DIANNA project aims at determining the best XAI methods in the context of scientific usage providing their OSS implementation based on the ONNX standard and demonstrations on benchmark datasets for images, text, time-series, and tabular data.

This page presents the output of three projects: DIANNA, funded by the Netherlands eScience Center and SUFR Alliance call 2020, the DIANNA+ Knowledge Development Project 2022-2023 and the Software Sustainability Project DIANNA meets the gods: Terra, Jupiter and Uriania. XAI for Earth, climate, and astronomy 2023-2024, both funded by the Netherlands eScience Center.

Participating organisations

Netherlands eScience Center
SURF

Impact

Output

How to find your Artificial Intelligence explainer

Author(s): Elena Ranguelova
Published in 2022

Team

Elena Ranguelova
Elena Ranguelova
Project leader
Netherlands eScience Center
Yang Liu
Yang Liu
lead RSE
Netherlands eScience Center
Leon Oostrum
Leon Oostrum
eScience Research Engineer
Netherlands eScience Center
Laura Ootes
Laura Ootes
Developer dashboard
Netherlands eScience Center
Pranav Chandramouli
Pranav Chandramouli
Fakhereh Alidoost
Fakhereh Alidoost
GC
Giulia Crocioni
developer dashboard
Netherlands eScience Center
Rena Bakhshi
Rena Bakhshi
Programme Manager
Netherlands eScience Center

Related projects

Explainable AI for Geo

XAI4GEO

Updated 10 months ago
In progress

AI4S2S

A high-level python package integrating expert knowledge and artificial intelligence to boost (sub) seasonal forecasting

Updated 18 months ago
In progress

Explainable Embeddings

What is happening in your machine-learned embedded spaces?

Updated 2 months ago
Finished

The nature of EU rules: strict and detailed, or lacking bite?

Updated 15 months ago
Finished

Recognising symbolism in Turkish television drama

Automated recognition of symbols in state-supported Turkish television series

Updated 1 month ago
Finished

Understanding visually grounded spoken language via multi-tasking

An alternative approach for intelligent systems to understand human speech

Updated 14 months ago
Finished

eScience Technology to Boost Quantum Dot Energy Conversion

More efficient lighting and solar energy conversion devices

Updated 18 months ago
Finished

Related software

DIANNA

DI

Deep Insight And Neural Network Analysis, DIANNA is the only Explainable AI, XAI library for scientists supporting Open Neural Network Exchange, ONNX - the de facto standard models format.

Updated 1 month ago
18 12

Explainable embeddings

EX

Experiments with regard to explanation of embedded spaces and multi modal models.

Updated 19 months ago
2