克服统计冲突并实现持续学习的定向结构适应
Directed Structural Adaptation to Overcome Statistical Conflicts and Enable Continual Learning
AI总结：
针对固定拓扑网络的统计冲突与灾难性遗忘问题，提出可定向增构的 DIRAD 及无需任务标签的 PREVAL 框架，实现高性能、简化网络和持续任务适应。
摘要：
如今的自适应网络依赖于过参数化的固定拓扑，无法突破其在所接触数据中遇到的统计冲突，并且当网络尝试复用现有结构学习新任务时，容易出现“灾难性遗忘”。我们提出一种结构适应方法 DIRAD，它能够根据需要以定向方式增加复杂度，而不受数据集内部统计冲突的限制。随后，我们扩展该方法并提出 PREVAL 框架，旨在通过检测新数据并将遇到的数据分配给适合处理这些数据的模型，来防止持续学习中的“灾难性遗忘”，且在整个工作流程中的任何环节都不需要任务标签。我们展示了 DIRAD 在扩展网络方面的可靠性：它能保持高性能，并且比固定拓扑网络简单数个数量级；同时论证了 PREVAL 的概念验证运行，其中可观察到对新任务的持续适应，并且系统能够检测和辨别先前遇到过的任务。
英文摘要：
Adaptive networks today rely on overparameterized fixed topologies that cannot break through the statistical conflicts they encounter in the data they are exposed to, and are prone to "catastrophic forgetting" as the network attempts to reuse the existing structures to learn new task. We propose a structural adaptation method, DIRAD, that can complexify as needed and in a directed manner without being limited by statistical conflicts within a dataset. We then extend this method and present the PREVAL framework, designed to prevent "catastrophic forgetting" in continual learning by detection of new data and assigning encountered data to suitable models adapted to process them, without needing task labels anywhere in the workflow. We show the reliability of the DIRAD in growing a network with high performance and orders-of-magnitude simpler than fixed topology networks; and demonstrate the proof-of-concept operation of PREVAL, in which continual adaptation to new tasks is observed while being able to detect and discern previously-encountered tasks.
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。
