Learning Conflict-Noticed Architecture for Multi-Task Learning

Link
Abstract

Multi-task learning has been widely used in many applications to enable more efficient learning by sharing part of the architecture across multiple tasks. However, a major challenge is the gradient conflict when optimizing the shared parameters, where the gradients of different tasks could have opposite directions. Directly averaging those gradients will impair the performance of some tasks and cause negative transfer. Different from most existing works that manipulate gradients to mitigate the gradient conflict, in this paper, we address this problem from the perspective of architecture learning and propose a Conflict-Noticed Architecture Learning (CoNAL) method to alleviate the gradient conflict by learning architectures. By introducing purely-specific modules specific to each task in the search space, the CoNAL method can automatically learn when to switch to purely-specific modules in the tree-structured network architectures when the gradient conflict occurs. To handle multi-task problems with a large number of tasks, we propose a progressive extension of the CoNAL method. Extensive experiments on computer vision, natural language processing, and reinforcement learning benchmarks demonstrate the effectiveness of the proposed methods.

Synth

Problem:: 다중 태스크 학습(MTL)에서 공유 파라미터의 Gradient 충돌 문제 발생/기존의 Gradient 조작 방법으로는 근본적인 아키텍처 충돌 해결 어려움 존재

Solution:: Purely-Specific Module을 도입해 태스크별로 충돌 시 독립적인 모듈 학습/동적 분기 시점을 자동 학습하여 최적의 아키텍처를 탐색

Novelty:: Gradient 충돌 문제를 아키텍처 설계 관점에서 접근한 최초의 시도

Note:: Task 충돌 문제를 GazeTarget의 Detection과 Estimation에 적용하려고 봤는데, 아키텍쳐적 접근 방식이라서 참고할게 없었음

Summary

Motivation

Method

file-20250327012344305.png

(a) LTB, (b) CoNAL, (c) CoNAL-Pro

Method 검증