Neural Representations in Multi-Task Learning guided by Task-Dependent Contexts

Santiago Galella, Salva Ardid

Keywords: Learning Representations, Neural Geometry, Context-Dependent Decision Making, Attention Mechanisms

TL;DR: We investigate neural representations learned by multi-task architectures focusing on task-switching networks that use task-dependent contexts.

Abstract:

The ability to switch between tasks effectively in response to external stimuli is a hallmark of cognitive control. Our brain is able to filter and to integrate external information to accomplish goal-directed behavior. Task switching occurs rapidly and efficiently, allowing us to perform multiple tasks with ease. In a similar way, deep learning models can be tailored to exhibit multi-task capabilities and achieve high performance across domains. Still, understanding how neural networks make predictions is crucial in many real-world applications.

In this study, we delve into neural representations learned by multi-tasking architectures. Concretely, we compare individual and parallel networks with task switching networks. Task-switching networks leverage task-dependent contexts to learn disentangled representations without hurting the overall task accuracy. We show that task-switching networks operate in an intermediate regime between individual and parallel. In addition, we show that shared representations are produced by the emergence neurons encoding multiple tasks. Furthermore, we study the role of contexts across network processing and show its role at aligning the task with the relevant features. Finally, we investigate how the magnitude of contexts affects the performance in task-switching networks.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Deep Learning and representational learning