受认知启发的增量漂移概念学习
Cognitively Inspired Learning of Incremental Drifting Concepts
AI总结：
针对持续学习中的分布漂移与灾难性遗忘，提出融合 PDP 与 CLS 理论的计算模型，通过多模态嵌入和伪数据回放实现新概念的增量学习。
摘要：
人类能够不断将已学知识扩展到新领域，并学习新概念，而不会对过去的学习经验产生任何干扰。相比之下，在输入数据分布随时间变化的持续学习场景中，机器学习模型表现不佳。受神经系统学习机制的启发，我们开发了一种计算模型，使深度神经网络能够在持续学习场景中学习新概念，并将其已学知识增量扩展到新领域。我们依托 Parallel Distributed Processing（并行分布式处理）理论，在嵌入空间中以多模态分布的形式编码抽象概念。该嵌入空间由网络隐藏层中的内部数据表示建模。我们还利用 Complementary Learning Systems（互补学习系统）理论，通过实现伪排练，为模型配备记忆机制以克服灾难性遗忘。我们的模型能够生成用于经验回放的伪数据点，并将新经验积累到过去已学经验中，而不会造成跨任务干扰。
英文摘要：
Humans continually expand their learned knowledge to new domains and learn new concepts without any interference with past learned experiences. In contrast, machine learning models perform poorly in a continual learning setting, where input data distribution changes over time. Inspired by the nervous system learning mechanisms, we develop a computational model that enables a deep neural network to learn new concepts and expand its learned knowledge to new domains incrementally in a continual learning setting. We rely on the Parallel Distributed Processing theory to encode abstract concepts in an embedding space in terms of a multimodal distribution. This embedding space is modeled by internal data representations in a hidden network layer. We also leverage the Complementary Learning Systems theory to equip the model with a memory mechanism to overcome catastrophic forgetting through implementing pseudo-rehearsal. Our model can generate pseudo-data points for experience replay and accumulate new experiences to past learned experiences without causing cross-task interference.
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。
