AdapMTL：面向多任务学习模型的自适应剪枝框架
AdapMTL: Adaptive Pruning Framework for Multitask Learning Model
AI总结：
本文提出 AdapMTL 自适应剪枝框架，通过共享主干与任务头的可学习软阈值及任务损失自适应加权，在 NYU-v2 和 Tiny-Taskonomy 上实现优于现有方法的多任务精度与稀疏性平衡。
摘要：
在多媒体与多模态处理领域，高效处理图像、视频和传感器数据等多种数据流至关重要。模型压缩与多任务学习（MTL）在该领域十分关键，为满足同时处理和解释多种媒体形式的资源密集型需求提供了可能。然而，由于需要在多个任务之间平衡稀疏性分配与精度性能，有效压缩多任务模型面临重大挑战。为应对这些挑战，我们提出了 AdapMTL，这是一种面向 MTL 模型的自适应剪枝框架。AdapMTL 利用多个可学习的软阈值，分别独立分配给共享主干和任务特定头，以捕捉不同组件对剪枝敏感性的细微差异。在训练过程中，该框架联合优化软阈值和 MTL 模型权重，从而自动确定每个组件的合适稀疏度，以同时实现较高的任务精度和整体稀疏性。它还引入了自适应加权机制，根据每个任务对剪枝的鲁棒性动态调整任务特定损失的重要性。我们通过在常用多任务数据集 NYU-v2 和 Tiny-Taskonomy 上使用不同架构进行全面实验，证明了 AdapMTL 的有效性；与最先进的剪枝方法相比，其展现出更优越的性能。
英文摘要：
In the domain of multimedia and multimodal processing, the efficient handling of diverse data streams such as images, video, and sensor data is paramount. Model compression and multitask learning (MTL) are crucial in this field, offering the potential to address the resource-intensive demands of processing and interpreting multiple forms of media simultaneously. However, effectively compressing a multitask model presents significant challenges due to the complexities of balancing sparsity allocation and accuracy performance across multiple tasks. To tackle these challenges, we propose AdapMTL, an adaptive pruning framework for MTL models. AdapMTL leverages multiple learnable soft thresholds independently assigned to the shared backbone and the task-specific heads to capture the nuances in different components' sensitivity to pruning. During training, it co-optimizes the soft thresholds and MTL model weights to automatically determine the suitable sparsity level at each component to achieve both high task accuracy and high overall sparsity. It further incorporates an adaptive weighting mechanism that dynamically adjusts the importance of task-specific losses based on each task's robustness to pruning. We demonstrate the effectiveness of AdapMTL through comprehensive experiments on popular multitask datasets, namely NYU-v2 and Tiny-Taskonomy, with different architectures, showcasing superior performance compared to state-of-the-art pruning methods.
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。
