Website Optimization

Bandit Algorithms: epsilon-Greedy Algorithm

A/B testing can be defined as a randomized controlled experiment that allows us to test if there is a causal relationship between a change to a website/app and the user behavior. The change can be visible such as location of a button on the homepage or invisible such as the ranking/recommendation algorithms and backend infrastructure. Web/Mobile developers and business stakeholders always face the following dilemma: Should we try out all ideas and explore all options continuously?