AutoSeM: Automatic Task Selection and Mixing in Multi-Task Learning
arXiv:1904.04153
Abstract
Multi-task learning (MTL) has achieved success over a wide range of problems, where the goal is to improve the performance of a primary task using a set of relevant auxiliary tasks. However, when the usefulness of the auxiliary tasks w.r.t. the primary task is not known a priori, the success of MTL models depends on the correct choice of these auxiliary tasks and also a balanced mixing ratio of these tasks during alternate training. These two problems could be resolved via manual intuition or hyper-parameter tuning over all combinatorial task choices, but this introduces inductive bias or is not scalable when the number of candidate auxiliary tasks is very large. To address these issues, we present AutoSeM, a two-stage MTL pipeline, where the first stage automatically selects the most useful auxiliary tasks via a Beta-Bernoulli multi-armed bandit with Thompson Sampling, and the second stage learns the training mixing ratio of these selected auxiliary tasks via a Gaussian Process based Bayesian optimization framework. We conduct several MTL experiments on the GLUE language understanding tasks, and show that our AutoSeM framework can successfully find relevant auxiliary tasks and automatically learn their mixing ratio, achieving significant performance boosts on several primary tasks. Finally, we present ablations for each stage of AutoSeM and analyze the learned auxiliary task choices.
NAACL 2019 (12 pages)
References in corpus (6)
- Practical Bayesian Optimization of Machine Learning Algorithms
- A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning
- One Model To Learn Them All
- GPflowOpt: A Bayesian Optimization Library using TensorFlow
- Taming Non-stationary Bandits: A Bayesian Approach
- Identifying beneficial task relations for multi-task learning in deep neural networks