论文标题
学习为多任务学习分支
Learning to Branch for Multi-Task Learning
论文作者
论文摘要
通过共享网络的某些层，在一个深层网络中共同培训多个任务可以减少推理期间的延迟，并在单个任务中提高性能。但是，超出网络可能会错误地执行过度概括，从而导致跨任务的负面知识转移。先前的工作依赖于人类直觉或预先计算的任务相关性得分来进行临时分支结构。他们提供了最佳的最终结果，并且通常需要巨大的努力来进行试验过程。在这项工作中，我们提出了一种自动化的多任务学习算法，该算法学习在网络中在哪里共享或分支，设计了一个有效的网络拓扑，该网络拓扑直接针对跨任务的多个目标进行了优化。具体而言，我们提出了一个新颖的树结构设计空间，该空间将树枝分支操作作为gumbel-softmax采样过程。这可以实现可端到端训练的可区分网络分裂。我们验证了对受控的合成数据，Celeba和Taskomony的建议方法。
Training multiple tasks jointly in one deep network yields reduced latency during inference and better performance over the single-task counterpart by sharing certain layers of a network. However, over-sharing a network could erroneously enforce over-generalization, causing negative knowledge transfer across tasks. Prior works rely on human intuition or pre-computed task relatedness scores for ad hoc branching structures. They provide sub-optimal end results and often require huge efforts for the trial-and-error process. In this work, we present an automated multi-task learning algorithm that learns where to share or branch within a network, designing an effective network topology that is directly optimized for multiple objectives across tasks. Specifically, we propose a novel tree-structured design space that casts a tree branching operation as a gumbel-softmax sampling procedure. This enables differentiable network splitting that is end-to-end trainable. We validate the proposed method on controlled synthetic data, CelebA, and Taskonomy.
