There are numerous probable answers to the transparency issue. SHAP attempted to resolve the transparency challenges by visualising the contribution of each function for the output.[191] LIME can locally approximate a model with a less complicated, interpretable design.[192] Multitask learning offers a lot of outputs in addition to the concentrate