One Configuration to Rule Them All? Towards Hyperparameter Transfer in Topic Models using Multi-Objective Bayesian Optimization
This work addresses hyperparameter tuning for topic modeling, which is incremental as it applies existing optimization methods to a specific domain.
The paper tackled the problem of hyperparameter selection in topic models by conducting multi-objective Bayesian optimization across three models, revealing that optimal configurations depend on corpus characteristics and can be transferred between datasets.
Topic models are statistical methods that extract underlying topics from document collections. When performing topic modeling, a user usually desires topics that are coherent, diverse between each other, and that constitute good document representations for downstream tasks (e.g. document classification). In this paper, we conduct a multi-objective hyperparameter optimization of three well-known topic models. The obtained results reveal the conflicting nature of different objectives and that the training corpus characteristics are crucial for the hyperparameter selection, suggesting that it is possible to transfer the optimal hyperparameter configurations between datasets.