Challenges in Big Data and Traditional AI
In this introductory chapter, why federated learning (FL) is going to be a key technology in the 2020s is explained in detail. You will learn what big data is and how it has been problematic from the perspectives of data privacy, model bias, and drift. A solid understanding of the necessity of such issues and solutions for them will motivate you to embark on a challenging journey to acquire relevant knowledge and skills, using the following chapters to chart the mastery of FL. After reading this chapter, it will become obvious that there is a massive paradigm shift in artificial intelligence (AI) and machine learning (ML), which is happening due to public and business concerns over the current reliance on big data-oriented systems. Without further ado, let us depart!
In this chapter, we will cover the following topics:
- Understanding the nature of big data
- Data privacy as a bottleneck
- Impacts of training data and model bias
- Model drift and performance degradation
- FL as the main solution for data problems
