利用外部记忆持续学习多模态动力学
Continual Learning of Multi-modal Dynamics with External Memory
AI总结：
针对新模式依次出现且训练序列真实模式未知的多模态动力学问题，本文提出基于神经情景记忆和 Dirichlet Process 先验的描述符检索与迁移方法，持续学习性能优于主流参数迁移方法。
摘要：
我们研究了当新的行为模式依次出现时，将模型拟合到动力学环境的问题。学习模型能够感知新模式何时出现，但无法获取各条训练序列的真实模式。当前最先进的持续学习方法无法处理这一设定，因为参数迁移会遭受灾难性干扰，而情景记忆的设计又需要知道序列的真实模式。我们设计了一种新颖的持续学习方法，通过在神经情景记忆中维护所遇到序列模式的 descriptor（描述符），克服了上述两个限制。我们在记忆的注意力权重上采用 Dirichlet Process（狄利克雷过程）先验，以促进模式描述符的高效存储。我们的方法实现持续学习的方式是：在任务之间迁移知识，即从过去任务中检索与当前序列模式相似的模式描述符，并将该描述符作为控制输入送入其转移核。我们观察到，与主流参数迁移方法相比，我们的方法取得了更优的持续学习性能。
英文摘要：
We study the problem of fitting a model to a dynamical environment when new modes of behavior emerge sequentially. The learning model is aware when a new mode appears, but it cannot access the true modes of individual training sequences. The state-of-the-art continual learning approaches cannot handle this setup, because parameter transfer suffers from catastrophic interference and episodic memory design requires the knowledge of the ground-truth modes of sequences. We devise a novel continual learning method that overcomes both limitations by maintaining a \textit{descriptor} of the mode of an encountered sequence in a neural episodic memory. We employ a Dirichlet Process prior on the attention weights of the memory to foster efficient storage of the mode descriptors. Our method performs continual learning by transferring knowledge across tasks by retrieving the descriptors of similar modes of past tasks to the mode of a current sequence and feeding this descriptor into its transition kernel as control input. We observe the continual learning performance of our method to compare favorably to the mainstream parameter transfer approach.
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。
