Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts

Abstract

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 tradeo s between task-speci c 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 di erent levels of task relatedness, we rst 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 bene t, depending on di erent levels of randomness in the training data and model initialization. Furthermore, we demonstrate the performance improvements by MMoE on real tasks including a binary classi cation benchmark, and a large-scale content recommendation system at Google.

Fairness in Recommendation Ranking through Pairwise Comparisons

“…Here we evaluate the ranking model's performance. The ranking model is a multi-layer neural network trained in a pointwise-fashion with multiple heads to predict the probability of a click, y, and a set of user engagement signals after the click, which we refer to in the aggregate by z; this is a similar setup to [25,36]. This model is continuously trained on a dataset of interactions with previous recommendations.…”

“…These systems often follow cascading patterns with sequences of models being used [47,24]. More recently, there has been a strong growth in using state-of-the-art neural network techniques to improve recommender accuracy [25,16,36].…”

Towards Neural Mixture Recommender for Long Range Dependent User Sequences

“…Gating Network Borrowing the idea from from Mixture-of-Experts model [23,35], we build a gating network to aggregate our encoders' results. The gate is also helpful to better understand our 1 It is unnecessary if d in is same as d enc .…”

Meta Reinforcement Learning with Task Embedding and Shared Policy

“…A major limitation of most meta-RL methods (discussed thoroughly in § 2) is that they do not explicitly and adequately model the individuality and the commonness of tasks, which has proven to play an important role in the literature of multitask learning [Ruder, 2017;Ma et al, 2018] and should be likewise applicable to meta-RL. Take the case of locomotion tasks, where an agent needs to move to different target locations for different tasks.…”