In the information age of 21st century, data driven AI/ML has ushered a new era in plethora of industries such as medical, manufacturing, production and retail by helping stakeholders to take informed decisions. One shortcoming of data driven AI is that it functions primarily on statistical concept of correlation, which is far from causation (the fundamental principle behind human thinking). Thus, if we want AI to emulate human intelligence it is necessary to embed it with the causation principle. This can be achieved by Bayesian Networks (BNs) a primary tool for probabilistic reasoning under uncertainty and causal inference. A small introduction to BNs along with the illustrative case study shall be presented.
- Current data driven ML is far from Human Level Intelligence
- Causal inference is missing in current data driven ML techniques
- Bayesian Network, a tool for causal inference can abridge the gap between AI and Human Level Intelligence