Soumik Sarkar:
Role of Interpretable Machine Learning in Cyber-Agricultural Systems (Invited Talk)

Dr. Soumik Sarkar received his Ph.D. in Mechanical Engineering from Penn State in 2011. Currently he is an Associate Professor of Mechanical Engineering and Computer Science, a Walter W Wilson Faculty Fellow in Engineering and the Director of the Translational AI Center at Iowa State. Dr. Sarkar’s research interests include Machine Learning and Decision & Control with applications to Cyber-Physical Systems such as energy, transportation, design & manufacturing and agriculture systems. He co‐authored more than 200 peer-reviewed publications and received about $30M research funding over the past seven years. Dr. Sarkar is a recipient of the prestigious Young Investigator award from the US Air Force Office of Scientific Research (AFOSR) in 2017 and the NSF CAREER award in 2019. 

One of the grand challenges of our generation is to get ready to feed 9 billion people by 2050 with sustainable use of water and chemicals. However, our current agricultural system is not prepared for it. We are facing unprecedented challenges in adopting sustainable management practices, increasing production, and coping with pest and climate stressors that threaten yield; while running a profitable farm operation. With changing climate, our crops are facing deadly biotic and abiotic stresses and diseases. However, to tackle this we use chemicals indiscriminately, most of which goes to our rivers and water systems, creating dead zones in the river downstream. The harmful chemical run-off increases due to soil compaction which is a direct consequence of using heavy equipment.
In this talk, I will discuss our vision of a new cyber-agricultural system that leads to an ultra-precision technology to monitor plants or small plots individually and treat them with minimum amount of chemicals. In turn, this will lower the barrier to entry into agriculture, increase safety, minimize runoff as well as soil compaction. To realize this grand vision, machine learning (ML) will play a large role. However, several key aspects of ML such as robustness, interpretability and data requirement need to be studied in the context of agriculture for successful deployment. In this regard, I will discuss a few success stories about interpretable ML in agricultural applications.