Soup-of-Experts: Pretraining Specialist Models via Parameters Averaging
This addresses the need for rapid deployment of multiple specialist models under size constraints, though it is incremental in leveraging parameter averaging.
The paper tackles the problem of efficiently creating specialized models for different data domain mixtures by proposing Soup-of-Experts, an architecture that instantiates models at test time for any domain weights without retraining, achieving quick deployment of small specialist models on language tasks.
Machine learning models are routinely trained on a mixture of different data domains. Different domain weights yield very different downstream performances. We propose the Soup-of-Experts, a novel architecture that can instantiate a model at test time for any domain weights with minimal computational cost and without re-training the model. Our architecture consists of a bank of expert parameters, which are linearly combined to instantiate one model. We learn the linear combination coefficients as a function of the input domain weights. To train this architecture, we sample random domain weights, instantiate the corresponding model, and backprop through one batch of data sampled with these domain weights. We demonstrate how our approach obtains small specialized models on several language modeling tasks quickly. Soup-of-Experts are particularly appealing when one needs to ship many different specialist models quickly under a model size constraint.