arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.05092cs.LGcs.CV
收藏

一种树型结构的多任务模型推荐器

A Tree-Structured Multi-Task Model Recommender

University of Massachusetts Amherst(马萨诸塞大学阿默斯特分校)

Lijun Zhang, Xiao Liu, Hui Guan

摘要:

树结构的多任务架构已被用于在多任务学习(MTL)的背景下联合处理多个视觉任务。主要的挑战是确定在给定骨干模型的情况下,每个任务在哪里分支,以优化任务精度和计算效率。为了解决这一挑战,本文提出了一个推荐器,给定一组任务和一个基于卷积神经网络的骨干模型,自动建议树结构的多任务架构,可以实现高的任务性能,同时满足用户指定的计算预算,而无需执行模型训练。广泛的评估流行的MTL基准测试表明,推荐的架构可以实现竞争力的任务精度和计算效率相比,最先进的MTL方法。我们的树结构多任务模型推荐器是开源的,可在https://github.com/zhanglijun95/TreeMTL上获得。

英文摘要:

Tree-structured multi-task architectures have been employed to jointly tackle multiple vision tasks in the context of multi-task learning (MTL). The major challenge is to determine where to branch out for each task given a backbone model to optimize for both task accuracy and computation efficiency. To address the challenge, this paper proposes a recommender that, given a set of tasks and a convolutional neural network-based backbone model, automatically suggests tree-structured multi-task architectures that could achieve a high task performance while meeting a user-specified computation budget without performing model training. Extensive evaluations on popular MTL benchmarks show that the recommended architectures could achieve competitive task accuracy and computation efficiency compared with state-of-the-art MTL methods. Our tree-structured multi-task model recommender is open-sourced and available at https://github.com/zhanglijun95/TreeMTL.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑