vix.ing · top · new · best · stats · spec

HyperGrid: Efficient Multi-Task Transformers with Grid-wise Decomposable Hyper Projections

2020/07/12 by Tay, Yi, Zhao, Zhe, Bahri, Dara +2
#Computation and Language (cs.CL) #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.2007.05891

Abstract

Achieving state-of-the-art performance on natural language understanding tasks typically relies on fine-tuning a fresh model for every task. Consequently, this approach leads to a higher overall parameter cost, along with higher technical maintenance for serving multiple models. Learning a single multi-task model that is able to do well for all the tasks has been a challenging and yet attractive proposition. In this paper, we propose HyperGrid, a new approach for highly effective multi-task learning. The proposed approach is based on a decomposable hypernetwork that learns grid-wise projections that help to specialize regions in weight matrices for different tasks. In order to construct the proposed hypernetwork, our method learns the interactions and composition between a global (task-agnostic) state and a local task-specific state. We apply our proposed HyperGrid on the current state-of-the-art T5 model, demonstrating strong performance across the GLUE and SuperGLUE benchmarks when using only a single multi-task model. Our method helps bridge the gap between fine-tuning and multi-task learning approaches.

Related