Exploring Shared Structures and Hierarchies for Multiple NLP Tasks
arXiv:1808.07658
Abstract
Designing shared neural architecture plays an important role in multi-task learning. The challenge is that finding an optimal sharing scheme heavily relies on the expert knowledge and is not scalable to a large number of diverse tasks. Inspired by the promising work of neural architecture search (NAS), we apply reinforcement learning to automatically find possible shared architecture for multi-task learning. Specifically, we use a controller to select from a set of shareable modules and assemble a task-specific architecture, and repeat the same procedure for other tasks. The controller is trained with reinforcement learning to maximize the expected accuracies for all tasks. We conduct extensive experiments on two types of tasks, text classification and sequence labeling, which demonstrate the benefits of our approach.
arXiv comment8
References in corpus (8)
- Adam: A Method for Stochastic Optimization
- Neural Architecture Search with Reinforcement Learning
- Seeing stars: Exploiting class relationships for sentiment categorization with respect to rating scales
- Efficient Neural Architecture Search via Parameter Sharing
- Hierarchical Representations for Efficient Architecture Search
- Reward Augmented Maximum Likelihood for Neural Structured Prediction
- Routing Networks: Adaptive Selection of Non-linear Functions for Multi-Task Learning
- Beyond Shared Hierarchies: Deep Multitask Learning through Soft Layer Ordering
Cited by in corpus (6)
- AutoML: A Survey of the State-of-the-Art
- Multi-Task Learning with Deep Neural Networks: A Survey
- AdaShare: Learning What To Share For Efficient Deep Multi-Task Learning
- Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap
- Meta-Learning Multi-task Communication
- TimeAutoML: Autonomous Representation Learning for Multivariate Irregularly Sampled Time Series