arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.00115cs.LGcs.AI
收藏

实例温度知识蒸馏

Instance Temperature Knowledge Distillation

Singapore University of Technology and Design(新加坡科技设计大学);Wuhan University(武汉大学)

Zhengbo Zhang, Yuxi Zhou, Jia Gong, Jun Liu, Zhigang Tu

AI总结:

实例温度知识蒸馏通过强化学习方法优化温度调整,提升知识蒸馏的性能和效率。

摘要:

知识蒸馏(KD)通过允许学生网络逐步学习教师网络传递的知识来提升其性能。现有方法动态调整温度以使学生网络适应KD不同学习阶段的变动学习难度。KD是一个连续过程,但在调整温度时,这些方法只考虑当前学习阶段的操作即时收益,而未考虑其未来的回报。为了解决这个问题,我们将温度调整视为一个连续决策任务,并提出基于强化学习的方法,称为RLKD。重要的是,我们设计了一种新的状态表示,使智能体能够做出更有信息的行动(即实例温度调整)。为了解决由于KD设置导致的方法中延迟奖励问题,我们探索了实例奖励校准方法。此外,我们设计了一种高效的探索策略,使智能体能够更高效地学习有价值的实例温度调整策略。我们的框架可以作为插件技术轻松插入各种KD方法中,我们在图像分类和目标检测任务上验证了其有效性。我们的项目地址是https://itkd123.github.io/ITKD.github.io。

英文摘要:

Knowledge distillation (KD) enhances the performance of a student network by allowing it to learn the knowledge transferred from a teacher network incrementally. Existing methods dynamically adjust the temperature to enable the student network to adapt to the varying learning difficulties at different learning stages of KD. KD is a continuous process, but when adjusting the temperature, these methods consider only the immediate benefits of the operation in the current learning phase and fail to take into account its future returns. To address this issue, we formulate the adjustment of temperature as a sequential decision-making task and propose a method based on reinforcement learning, termed RLKD. Importantly, we design a novel state representation to enable the agent to make more informed action (i.e. instance temperature adjustment). To handle the problem of delayed rewards in our method due to the KD setting, we explore an instance reward calibration approach. In addition,we devise an efficient exploration strategy that enables the agent to learn valuable instance temperature adjustment policy more efficiently. Our framework can serve as a plug-and-play technique to be inserted into various KD methods easily, and we validate its effectiveness on both image classification and object detection tasks. Our project is at https://itkd123.github.io/ITKD.github.io/.

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。

↑