As domains like healthcare look to deploy artificial intelligence and deep learning systems, where questions of accountability and transparency are particularly important, Explaining reason for decision is important aspect.
Current widely used ML algorithms doesn't explain reason. If you are interested to know about this important aspect, this document helps
Explainability is the extent to which the internal mechanics of a machine or deep learning system can be explained in human terms.
It doesn’t take a data scientist to work out that the machine and deep learning algorithms built into automation and artificial intelligence systems lack transparency.
Why explanation is needed?
A typical example is diagnosing medical conditions. ML models can be produced using data such as X-rays and CT scans, and then be fed with new data and asked to identify whether a medical condition is present or not. In situations like these, however, getting an outcome is not enough: we need to know the explanation behind it, and this is where it gets tricky.
Explainable AI behaviour
Please notice the reason for risk section in below picture. It is produced by the ML.
Through the introduction of probabilistic learning, applications became increasingly successful, but increasingly opaque.
The lack of understanding within Artificial Neural Networks has been a large barrier to the adoption of machine learning. This uncertainty comes from the “Black Box” of the neural net.
It is very important for the adoption of machine learning models and ANN’s to be incorporated into fields such as healthcare and finance.
Explainable AI deals with the implementation of transparency and traceability of statistical black‐box machine learning methods, particularly deep learning (DL).
It is emerging field. This paper evaluates explainable AI models
Interpretability and explainability are different terms
https://www.kdnuggets.com/2018/12/machine-learning-explainability-interpretability-ai.html
https://images.app.goo.gl/b2r6Pm6SVK5insAy6
https://images.app.goo.gl/cgzcoQ3xqnLhSqg38
https://images.app.goo.gl/ag4wp1hxeSkFTPhj6
https://onlinelibrary.wiley.com/doi/full/10.1002/widm.1312
https://medium.com/@shagunm1210/the-explainable-neural-network-8f95256dcddb
https://arxiv.org/abs/2005.01831
https://images.app.goo.gl/a4JNVxkbha1LioTSA