# Many Task Learning with Task Routing

## TL;DR

This paper introduces Task Routing, a method for multi-task learning that efficiently handles over 20 tasks simultaneously by applying conditional feature transformations, reducing complexity and resource needs.

## Contribution

The paper proposes Task Routing layers enabling large-scale multi-task learning with hundreds of tasks in a single model, improving efficiency and scalability.

## Key findings

- Successfully fits hundreds of classification tasks in one model.
- Outperforms strong baselines and state-of-the-art on 5 datasets.
- Reduces architectural complexity and resource requirements.

## Abstract

Typical multi-task learning (MTL) methods rely on architectural adjustments and a large trainable parameter set to jointly optimize over several tasks. However, when the number of tasks increases so do the complexity of the architectural adjustments and resource requirements. In this paper, we introduce a method which applies a conditional feature-wise transformation over the convolutional activations that enables a model to successfully perform a large number of tasks. To distinguish from regular MTL, we introduce Many Task Learning (MaTL) as a special case of MTL where more than 20 tasks are performed by a single model. Our method dubbed Task Routing (TR) is encapsulated in a layer we call the Task Routing Layer (TRL), which applied in an MaTL scenario successfully fits hundreds of classification tasks in one model. We evaluate our method on 5 datasets against strong baselines and state-of-the-art approaches.