CURATED AND UPDATED INFORMATION ON ALL FACETS OF XAI
Beyond XAI ?
https://www.ayasdi.com/beyond-explainability-ai-transparency/
Beyond Explainability – AI Transparency – AyasdiAI
It is now well understood that in order to make Artificial Intelligence broadly useful, it is critical that humans can interact with and have confidence in the algorithms that are being used. This observation has led to the development of the notion of explainable AI (sometimes called XAI).
Model Cards for Model Reporting
https://arxiv.org/pdf/1810.03993.pdf
GUIDE TO INTERPRETABLE MACHINE LEARNING
https://www.topbots.com/interpretable-machine-learning/
If you can’t explain it simply, you don’t understand it well enough. — Albert Einstein Disclaimer
Explainability vs interpretability
https://bdtechtalks.com/2020/07/27/black-box-ai-models/
Explainable anatomical shape
Spiral: Explainable anatomical shape analysis through deep hierarchical generative models
https://arxiv.org/abs/1907.00058
SPIRAL.IMPERIAL.AC.UK
Interpretable policy derivation for reinforcement learning based on evolutionary feature synthesis
LINK.SPRINGER.COM
https://link.springer.com/article/10.1007/s40747-020-00175-y
Reinforcement learning based on the deep neural network has attracted much attention and has been widely used in real-world applications. However, the black-box property limits its usage from applying in high-stake areas, such as manufacture and healthcare.
Self-explainable AI
https://bdtechtalks.com/2020/06/15/self-explainable-artificial-intelligence/
The case for self-explainable AI
Scientist Daniel Elton discusses why we need artificial intelligence models that can explain their decisions by themselves as humans do.
Evolution of Classifier Confusion on the Instance Level
ARXIV.ORG: https://arxiv.org/abs/2007.11353
Deep meta-learning XAI
Explainable Artificial Intelligence (xAI) Approaches and Deep Meta-Learning Models | IntechOpen
The explainable artificial intelligence (xAI) is one of the interesting issues that has emerged recently. Many researchers are trying to deal with the subject with different dimensions and interesting results that have come out.
Explaining Deep Neural Networks using Unsupervised Clustering
https://arxiv.org/pdf/2007.07477.pdf
ARXIV.ORG
Interactive Studio for Explanatory Model Analysis, This is an R package.
https://cran.r-project.org/web/packages/modelStudio/modelStudio.pdf
CRAN.R-PROJECT.ORG
Automated Reasoning for Explainable AI
http://kocoon.gforge.inria.fr/slides/marques-silva.pdf
KOCOON.GFORGE.INRIA.FR
The “first” XAI libraries fusion
https://blog.fiddler.ai/2020/07/fiddler-captum-collaborate-on-explainable-ai/
BLOG.FIDDLER.AI
Fiddler Labs
AI with trust, visibility, and insightts built in. Fiddler is a breakthrough AI engine with explainability at its heart.
AI perspective on understanding and meaning
BDTECHTALKS.COM
https://bdtechtalks.com/2020/07/13/ai-barrier-meaning-understanding/
AI’s struggle to reach “understanding” and “meaning”
Computer scientist Melanie Mitchell breaks down the key elements that could allow artificial intelligence algorithms to grasp the “meaning” of things.
Robust Decision Tree
https://link.springer.com/chapter/10.1007/978-3-030-50153-2_36
Robust Predictive-Reactive Scheduling: An Information-Based Decision Tree Model
LINK.SPRINGER.COM
In this paper we introduce a proactive-reactive approach to deal with uncertain scheduling problems.
Yellowbrick directly from Scikit
SCIKIT-YB.ORG
Yellowbrick: Machine Learning Visualization — Yellowbrick v1.3.post1 documentation (scikit-yb.org)
Yellowbrick extends the Scikit-Learn API to make model selection and hyperparameter tuning easier. Under the hood, it’s using Matplotlib.
Levels of XAI framework
https://link.springer.com/chapter/10.1007/978-3-030-51924-7_6
Decision Theory Meets Explainable AI. A CIU without the HAP !
https://link.springer.com/chapter/10.1007/978-3-030-51924-7_4
LINK.SPRINGER.COM
Decision Theory Meets Explainable AI
Explainability has been a core research topic in AI for decades and therefore it is surprising that the current concept of Explainable AI (XAI) seems to have been launched as late as 2016.
ExplainX.ai
https://github.com/explainX/explainx
GITHUB.COM
Explain any Black-Box Machine Learning Model with explainX: Fast, Scalable & State-of-the-art Explainable AI Platform. – explainX/explainx
Explainable 3D Convolutional Neural Networks by Learning Temporal Transformations
DEEPAI.ORG
In this paper we introduce the temporally factorized 3D convolution (3TConv) as an interpretable alternative to the regular 3D convolutions.
XAI package DALEX
https://github.com/ModelOriented/DALEX
GITHUB.COM
moDel Agnostic Language for Exploration and eXplanation – ModelOriented/DALEX
AIMLAI, Submission deadline: Jul 22, 2020
https://project.inria.fr/aimlai/
PROJECT.INRIA.FR
Advances in Interpretable Machine Learning and Artificial Intelligence (AIMLAI)
CONSAC: Robust Multi-Model Fitting by Conditional Sample Consensus
https://arxiv.org/pdf/2001.02643.pdf
also code available: https://github.com/fkluger/consac
fkluger/consac
CONSAC: Robust Multi-Model Fitting by Conditional Sample Consensus – fkluger/consac
The four dimensions of contestable AI diagnostics- A patient-centric approach to explainable AI
SCIENCEDIRECT.COM
https://www.sciencedirect.com/science/article/pii/S0933365720301330
The problem of the explainability of AI decision-making has attracted considerable attention in recent years.
When Explanations Lie
https://arxiv.org/abs/1912.09818
(code page not found on my computer !)
ARXIV.ORG
When Explanations Lie: Why Many Modified BP Attributions Fail
Attack to Explain Deep Representation
OPENACCESS.THECVF.COM
Funny title from Google: Neural Networks Are More Productive Teachers Than Human Raters 🙂, the paper is as you might expect related to knowledge distillation from a black box. It is accepted at CVPR taking place this week !
https://arxiv.org/pdf/2003.13960.pdf
ARXIV.ORG
InterpretML from Microsoft
https://github.com/interpretml/interpret
GITHUB.COM
Fit interpretable models. Explain blackbox machine learning. – interpretml/interpret
SK-MOEFS: A Library in Python for Designing Accurate and Explainable Fuzzy Models
LINK.SPRINGER.COM
SK-MOEFS: A Library in Python for Designing Accurate and Explainable Fuzzy Models
Recently, the explainability of Artificial Intelligence (AI) models and algorithms is becoming an important requirement in real-world applications.
Fooling LIME and SHAP
https://arxiv.org/abs/1911.02508
ARXIV.ORG
Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methods
Explainable cooperative machine learning
https://link.springer.com/article/10.1007/s13218-020-00632-3
LINK.SPRINGER.COM
eXplainable Cooperative Machine Learning with NOVA
In the following article, we introduce a novel workflow, which we subsume under the term “explainable cooperative machine learning” and show its practical application in a data annotation and model training tool called NOVA.
XAI research job in Rome
https://euraxess.ec.europa.eu/jobs/527048
Neural Graph Learning
STORAGE.GOOGLEAPIS.COM
If you want to play around, maybe earn some money:
https://www.innocentive.com/ar/challenge/browse?categoryName=Biology
INNOCENTIVE.COM
InnoCentive Challenge Center
LIMEtree>>>>> https://arxiv.org/pdf/2005.01427.pdf
Explainable AI Through Combination of Deep Tensor and Knowledge Graph
FUJITSU.COM
Master thesis in Quantifying the Performance of Explainability Algorithms, University of Waterloo, 2020
https://uwspace.uwaterloo.ca/bitstream/handle/10012/15922/Lin_ZhongQiu.pdf?sequence=5&isAllowed=y
UWSPACE.UWATERLOO.CA
XAI by Topological Hierarchical Decomposition
https://math.osu.edu/events/topology-geometry-and-data-seminar-ryan-kramer
also the paper: https://arxiv.org/abs/1811.10658
Topology, Geometry and Data Seminar – Ryan Kramer
MATH.OSU.EDU
A very simple manner to image XAI related to the way our brain thinks
https://hbr.org/2017/05/linear-thinking-in-a-nonlinear-world
XAI for COVID-19 classification, actually a RF
https://www.medrxiv.org/node/82227.external-links.html
MEDRXIV.ORG
An Interpretable Machine Learning Framework for Accurate Severe vs Non-severe COVID-19 Clinical Type Classification
Effectively and efficiently diagnosing COVID-19 patients with accurate clinical type is essential to achieve optimal outcomes of the patients as well as reducing the risk of overloading the healthcare system.
More Python XAI tools !
https://pyss3.readthedocs.io/en/latest/
Welcome to PySS3’s documentation! — PySS3 0.5.9 documentation
PYSS3.READTHEDOCS.IO
PySS3 is a Python package that allows you to work with The SS3 Classification Model in a very straightforward, interactive and visual way.
XAI Critics !
Interpreting Interpretability !
http://www-personal.umich.edu/~harmank/Papers/CHI2020_Interpretability.pdf
WWW-PERSONAL.UMICH.EDU
gshap 0.0.3, latest released !
https://pypi.org/project/gshap/
gshap from PYPI.ORG
A technique in explainable AI for answering broader questions in machine learning.
Machine learning-based XAI
https://ieeexplore.ieee.org/document/9007737
IEEEXPLORE.IEEE.ORG
Explainable Machine Learning for Scientific Insights and Discoveries – IEEE Journals & Magazine