Knowledge Infused Policy Gradients with Upper Confidence Bound for Relational Bandits
This work addresses the problem of improving exploration-exploitation efficiency in relational bandits for applications like recommendation systems and healthcare, though it appears incremental as it adapts existing methods to a new setting.
The authors tackled the challenge of using relational context representations in contextual bandits, which expands the context space and complicates exploration-exploitation, by proposing Knowledge Infused Policy Gradients and a novel Upper Confidence Bound algorithm. They demonstrated that expert knowledge can significantly reduce total regret in some real-life datasets, such as a simulated music recommendation system, but not in others.
Contextual Bandits find important use cases in various real-life scenarios such as online advertising, recommendation systems, healthcare, etc. However, most of the algorithms use flat feature vectors to represent context whereas, in the real world, there is a varying number of objects and relations among them to model in the context. For example, in a music recommendation system, the user context contains what music they listen to, which artists create this music, the artist albums, etc. Adding richer relational context representations also introduces a much larger context space making exploration-exploitation harder. To improve the efficiency of exploration-exploitation knowledge about the context can be infused to guide the exploration-exploitation strategy. Relational context representations allow a natural way for humans to specify knowledge owing to their descriptive nature. We propose an adaptation of Knowledge Infused Policy Gradients to the Contextual Bandit setting and a novel Knowledge Infused Policy Gradients Upper Confidence Bound algorithm and perform an experimental analysis of a simulated music recommendation dataset and various real-life datasets where expert knowledge can drastically reduce the total regret and where it cannot.