Abstract: Trust and transparency is a critical challenge for many organizations in adopting AI. AI systems are often opaque and frequently we don’t understand how they came to the recommendations they make and what assumptions that they’re based on . Building more explainable models, while maintaining a high level of accuracy and helping humans understand the “Why” can increase productivity, adoption and innovation. My goal is to introduce you to explainable AI and encourage you to incorporate explainability into your ML workflows. In this talk I will share several examples of applications and how explainable AI can enhance these applications.
Bio: Haritha Yanam is the Director of Data science, Innovation at Liberty Mutual Insurance, where she focuses on building AI/ML solutions which help in mitigating Insurance Risk. Before joining Liberty Haritha worked at several fortune 500 companies leading data science and data engineering teams. Haritha comes with a strong Data Science/Machine Learning, Data Engineering & Data Analytics background. She enjoys teaching and currently works as an adjunct professor(part-time) at the University of Maryland Baltimore teaching data science.