1月12日论文推荐（附下载地址）
论文
Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts
作者
Jiaqi Ma (University of Michigan, Ann Arbor)
Zhe Zhao (Google Inc.)
Xinyang Yi (Google Inc.)
Jilin Chen (Google Inc.)
Lichan Hong (Google Inc.)
Ed H. Chi (Google Inc.)
推荐理由
“Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts”是一篇基于神经网络的多任务学习模型。本质上，这是个多任务学习的扩展，多任务学习在很多实习系统中都有应用，比如推荐。基于神经网络的多任务学习本质上就是在多个任务之间加一个共享表示层，从数学上可以证明这个共享表示层可以起到正则化的效果，提高模型的泛化能力，最简单的共享层就是如下图(a)所示。本文是提出一个多层共享表示层的模型，如下图(c)，MoE（Mixed of Experts），也就是在多个任务之间学习多个共享层，然后mixture起来。在mixture的时候可以加上单gate和多gate的学习机制。
最后在UCI的数据集上，这个方法相比传统方法有一定的提升。下图是一个实验结果。
后来作者还在Google的大规模数据上进行了实验，也取得一定的提升。
摘要
Neural-based multi-task learning has been successfully used in many real-world large-scale applications such as recommendation systems. For example, in movie recommendations, beyond providing users movies which they tend to purchase and watch, the system might also optimize for users liking the movies afterwards. With multi-task learning, we aim to build a single model that learns these multiple goals and tasks simultaneously. However, the prediction quality of commonly used multi-task models is often sensitive to the relationships between tasks. It is therefore important to study the modeling tradeos between task-specific objectives and inter-task relationships.
In this work, we propose a novel multi-task learning approach, Multi-gate Mixture-of-Experts (MMoE), which explicitly learns to model task relationships from data.We adapt the Mixture-of Experts (MoE) structure to multi-task learning by sharing the expert submodels across all tasks, while also having a gating network trained to optimize each task.To validate our approach on data with different levels of task relatedness, we first apply it to a synthetic dataset where we control the task relatedness.We show that the proposed approach performs better than baseline methods when the tasks are less related. We also show that the MMoE structure results in an additional trainability benefit, depending on different levels of randomness in the training data and model initialization. Furthermore, we demonstrate the performance improvements by MMoE on real tasks including a binary classification benchmark, and a large-scale content recommendation system at Google.